Between Craft and Science: Technical Work in the United States 9781501720888

Citation preview

BETWEEN CRAFT

AND

SCIENCE

A VOLUME IN THE

COLLECTION ON

0 .~~.~~ ..

"'Ii"' . ~

~::

and

U·

ua·. WORK ~·

edited by STEPHEN R. BARLEY

BETWEEN CRAFT AND SCIENCE Technical Work in U.S. Settings EDITED BY

Stephen R. Barley and Julian E. Orr

ILR Press

I an imprint of

Cornell University Press Ithaca and London

Copyright © 1997 by Cornell University All rights reserved. Except for brief quotations in a review, this book, or parts thereof, must not be reproduced in any form without permission in writing from the publisher. For information, address Cornell University Press, Sage House, 512 East State Street, Ithaca, New York 14850 First published 1997 by Cornell University Press. First printing, Cornell Paperbacks, 1997.

Library of Congress Cataloging-in-Publication Data Between craft and science: technical work in U.S. settings I edited by Stephen R. Barley and Julian E. Orr. p. em.~ (Collection on technology and work) Includes index. ISBN 0-8014-3296-0 (cloth: alk. paper).~ ISBN 0-8014-8366-2 (pbk. : alk. paper) I. Industrial technicians~United States. I. Barley. Stephen R. II. Orr, Julian E. (Julian Edgerton), 1945III. Series. TAI58.B47 1997 609.2'273~DC20 96-32077

This book is printed on Lyons Falls Turin Book, a paper that is totally chlorine-free and acid-free.

Cloth printing

10

9

8

7

6

5

4

3

2

Paperback printing

10

9

8

7

6

5

4

3

2

CONTENTS

Preface vii List of Contributors ix

Introduction: The Neglected Workforce 1 Stephen R. Barley and Julian E. Orr

Part I

Technical Work's Challenge to the Established Order

Technical Work in the Division of Labor: Stalking the Wily Anomaly 23 Peter Whalley and Stephen R. Barley

2 Technical Dissonance: Conflicting Portraits of Technicians 53 Jeffrey Keefe and Denise Potosky

3 Whose Side Are They On? Technical Workers and Management Ideology 82 Sean Creighton and Randy Hodson

Part II

Studies ofTechnical Practice, Knowledge, and Culture

4 Cutting Up Skills: Estimating Difficulty as an Element of Surgical 101 and Other Abilities Trevor Pinch, H. M. Collins, and Larry Carbone

CONTENTS

VI

5 Bleeding Edge Epistemology: Practical Problem Solving in Software Support Hot Lines 113 Brian T. Pentland

6 Computers, Clients, and Expertise: Negotiating Technical Identities in a Nontechnical World 129 Stacia E. Zabusky

7 Work as a Moral Act: How Emergency Medical Technicians Understand Their Work 154 Bonalyn J. Nelsen Part Ill

Implications ofTechnical Practice for Training, Credential ling, and Careers

8 The Infamous "Lab Error": Education, Skill, and Quality in Medical Technicians' Work 187 Mario Scarselletta

9 Engineering Education and Engineering Practice: Improving the 210 Fit Louis L. Bucciarelli and Sarah Kuhn

I 0 The Senseless Submergence of Difference: Engineers, Their Work, 230 and Their Careers Leslie Perlow and Lotte Bailyn References 245 Index 257

PREFACE

Aside from the work of doctors, lawyers, and a handful of other visible professionals, relatively little is known about the content or the social organization of technical work. To be sure, science and engineering have attracted considerable attention over the years, but only recently have researchers begun to examine what scientists and engineers actually do and how their work is organized. Information on the work of technicians is even scarcer, although technicians now represent 3.6 percent of the American labor force and have been the fastest growing occupational category for several decades. The chapters in this book take a step toward remedying this situation. Books are usually accretions of personal agendas, and this one is no exception. After studying technicians' work in several settings, we became convinced that the implications of the so-called shift to a postindustrial or service economy could not be fully understood without an appreciation of the expanding role that technical workers play in modem organizations. We thought progress toward such an understanding would be enhanced if researchers interested in technical work had an opportunity to pool their knowledge and lay the foundations of a research community. Although a small but growing number of scholars had become interested in technical work by the late 1980s, our perception was that many were unaware of each other's research because they spanned disciplines as diverse as sociology, anthropology, psychology, economics, engineering, and labor relations. We decided to remedy this situation. In 1991, under the auspices of the Program on Technology and Work at Cornell University's School of Industrial and Labor Relations, we embarked on the events that produced this volume. In November 1992, with funding from the Department of Labor and Cornell's Institute for Labor Market Policy, we hosted a workshop at Cornell University on the technical labor force. The workshop brought together sixteen scholars

vm

PREFACE

from a variety of disciplines who had conducted field studies of technical work, but who were mostly unfamiliar with each other's findings. In addition to most of the authors whose chapters appear in this volume, Charles Goodwin, an anthropologist from the University of South Carolina, and Patricia Sachs, an anthropologist employed by NYNEX, also attended. The workshop's goals were to facilitate the sharing of information and ideas, to arrive at a working definition of technical work, to identify issues that all participants considered critical for future development, to facilitate collaboration, and, finally, to lay the foundation for this book and a second, somewhat larger, conference. During the workshop, the participants converged on a series of topics that they agreed deserved further elaboration. They then divided responsibility for developing papers on these topics. The commissioned papers clustered around four broad themes: (1) technical work's challenge to the established order, (2) detailed studies of technical work, knowledge, and practice, (3) detailed studies of technical workers' identities, values, and beliefs, and (4) training, credentialling, and careers in technical occupations. The goal was to develop papers that ranged coherently from the theoretical to the descriptive to the policy-oriented. This volume retains the organizational scheme developed at the workshop. Between December 1992 and March 1994 the authors conducted additional research and produced initial drafts. These drafts were delivered at Cornell in March 1994 during a conference sponsored by the Alfred P. Sloan Foundation. The conference brought the authors together for three days with an audience composed of other researchers who have studied technical work as well as leaders from industry, labor, and government who are especially knowledgeable about issues pertaining to technical work and the technical labor force. After the conference, authors revised their papers in light of the dialogue that occurred. This volume would not have been possible without the financial and intellectual support of Hirsch Cohen at the Alfred P. Sloan Foundation, Stephanie Swirsky of the Department of Labor's Employment and Training Administration, and Ronald Ehrenberg, Director of the Institute for Labor Market Policy. The researchers and organizers associated with Cornell's Program on Technology and Work are also deeply indebted to the U.S. Department of Education which, from the beginning, has funded our work on technicians through grants from the National Center for the Educational Quality of the Workforce. Thanks are also due to Stacia Zabusky and Margaret Gleason, who organized the first workshop, to Renee Edelman, who orchestrated the conference, and to Paula Wright and Roger Kovalchick, who helped transform the papers into a single, integrated document. STEPHEN R. BARLEY JULIAN E. ORR Palo Alto, California October 11, 1995

CONTRIBUTORS

Lotte Bailyn Sloan School of Management Massachusetts Institute of Technology 50 Memorial Drive · Cambridge, Mass. 02139

Stephen R. Barley Industrial Engineering and Engineering Management 340 Terman Hall Stanford University Stanford, Calif. 94305

Louis L. Bucciarelli School of Engineering 3-282 Massachusetts Institute of Technology 77 Massachusetts Avenue Cambridge, Mass. 02139

Larry Carbone School of Veterinary Medicine Cornell University Ithaca, N.Y. 14853

H. M. Collins Centre for the Study of Knowledge, Expertise and Science Dept. of Sociology and Social Policy University of Southampton Southampton S017 lBS U.K. Sean Creighton Department of Sociology Ballantine Hall, Room 751 Indiana University Bloomington, Ind. 47401 Randy Hodson Department of Sociology Ballantine Hall, Room 751 Indiana University Bloomington, Ind. 47401 Jeffrey Keefe School of Management and Labor Relations Rutgers University P.O. Box 231 Ryders Lane New Brunswick, N.J. 08903

x

CONTRIBUTORS

Sarah Kuhn

Trevor Pinch

Policy and Planning Department College of Management University of Massachusetts-Lowell One University Avenue Lowell, Mass. 01853

Science and Technology Studies 622 Clark Hall Cornell University Ithaca, N.Y. 14853

Denise Potosky Bonalyn J, Nelsen Johnson Graduate School of Management Cornell University Ithaca, N.Y. 14853

Graduate Program in Management The Pennsylvania State University Great Valley Campus 30 E. Swedesford Road Malvern, Penn. 19355

Mario Scarselletta Julian E. Orr Xerox PARC 3333 Coyote Hill Road Palo Alto, Calif. 94304

Corning Inc. Science Products Division Corning, N.Y. 14831

Peter Whalley Brian T. Pentland School of Labor and Industrial Relations 412 South Kedsie Hall Michigan State University East Lansing, Mich. 48824-1032

Department of Sociology and Anthropology Loyola University Chicago Lake Shore Campus 6525 North Sheridan Road Chicago, Ill. 60626

Stacia E. Zabusky Leslie Perlow University of Michigan Business School 701 Tappan Street Ann Arbor, Mich. 48109-1234

Associate in Research Institute of European Studies Cornell University Ithaca, N.Y. 14853

BETWEEN CRAFT AND SCIENCE

INTRODUCTION: THE NEGLECTED WORKFORCE Stephen R. Barley and Julian E. Orr

THE CHANGING NATURE OF WORK

Work forms the bedrock of all economic systems. When the nature and social organization of work change, so does the fabric of society. On this dictum, Marx, Weber, Durkheim, and Tonnies anchored their analyses of the social transformation we now call the Industrial Revolution and, in the process, gave intellectual direction to the fledgling field of sociology. The Industrial Revolution transformed work in Western society in two related but analytically distinct ways. First and most obvious, it occasioned a massive shift in what people did for a living. Over the course of the nineteenth and early twentieth centuries, the balance of employment in every Western nation shifted from agriculture to manufacturing. Possibly more critical for the course of Western culture, however, was the second type of change: the shift in how people did their work, or what Marx referred to as a change in the "mode of production." Most goods produced and consumed up to the late nineteenth century were not that different from those produced and consumed before the Industrial Revolution. What changed was how goods were made. Prior to the Industrial Revolution, textiles, plows, muskets, glassware, chairs, tables, and most other products were manufactured by hand with simple tools either in the home or in the shops of skilled artisans. The Industrial Revolution brought workers together in larger shops and factories, where propinquity allowed owners to divide tasks into constituent activities and assign those activities to individuals who performed them repeatedly. The advent of interchangeable parts and specialized machine tools furthered the progressive division of labor which, with time, became the dis-

2

BARLEY AND ORR

tinctive signature of industrialization. By the early decades of the twentieth century, the shift in the nature of work was so far-reaching that "job" had come to mean employment in a vertically struqured organization, and craft had been reduced to a secondary means of organizing work (Abbott 1989). In the late 1960s, a handful of sociologists began to argue from several perspectives that the West was once again embroiled in an economic transformation whose scope would rival that of the Industrial Revolution. Their claim rested, in part, on the observation that the labor force was becoming increasingly whitecollar. For Daniel Bell ( 1973) and other advocates of a "postindustrial economy" (Galbraith 1967; Touraine 1971 ), the transformation pivoted on the declining importance of manufacturing and the growing centrality of service industries that were heavily populated by "knowledge workers," whose elite consisted of scientists, technicians, and professionals. A number of French Marxists believed they saw in the same trends the rise of a "new working class." Because technical workers controlled the knowledge and technologies on which their expertise rested, Serge Mallet (1975) and Andre Gorz (1967) argued that the new working class would be more effective than the older working class at resisting the dynamics of capitalism. Still other sociologists argued that professional and technical workers actually represented a "new middle" rather than a "new working" class and that, therefore, their interests were largely aligned with the existing power structure (Poulantzas 1975; Wright 1978). All three schools ofthought met with considerable skepticism, in part because each oversimplified the complexity of the changes it observed and in part because each postulated a shift in culture or social structure that was based more on wishful thinking than on evidence. Although the postindustrialists and the neo-Marxists may have overinterpreted the changes they observed, with each passing year it becomes more and more difficult to deny the accuracy of their most central observation: the occupational division of labor is again changing, and it is changing in an apparently consistent direction. Trends in the United States since mid-century are illustrative. As Table 1.1 indicates, even after a century of steady decline, agriculture still employed 12 percent of all Americans in 1950. By 1991, however, agriculture employed a mere 3 percent of the labor force. Considering it alone, one might view the continuing decline in agricultural employment as little more than the completion of a trend begun in the early years of the Industrial Revolution, but concomitant developments suggest a different scenario. During the Industrial Revolution, dwindling agricultural employment was offset by the expansion of semi-skilled and unskilled blue-collar labor and, somewhat later, by the expansion of the clerical and administrative workforce. Since mid-century, however, the number of unskilled and semi-skilled jobs has steadily declined. Whereas 26 percent of all Americans worked as operatives and laborers in 1950, only 15 percent of the workforce were

3

INTRODUCTION

TABLE I. I Occupational categories as a percentage of the labor force, 1950-1991

Category

1950

Farmworkers Professionalffechnical Craft and kindred Operatives/Laborers Clerical and kindred Service Managerial/Administrative Sales workers

12% 8 14 26 12 II

9 7

1960

1970

1980

1991

6%

3% 14 14 23 18 13 8 7

3% 15 12 18 17 13 10

3% 17

10

14 24 15 12 8 7

II

II

15 16 14 13 12

Net Change

-9% 9 -3 -II

4 3 4 5

Note: Percentage employment by occupational category from 1950 to 1970 was calculated from employment data presented on page 139 of The Statistical History of the United States from Colonial Times to the Present (U.S. Bureau of the Census, 1976). Data for 1980 were taken from Klein's (1984) article which transforms 1980 data using the Census Bureau's category system developed in 1983. Data for 1991 are taken from the Statistical Abstract of the United States (U.S. Bureau of Commerce, 1991).

so employed by 1991. 1 Over the same period, the proportion of Americans employed in skilled crafts also declined from 14 to 11 percent. The demise of bluecollar work is hardly news. Less well recognized is that clerical work has also begun to wane. Clerical employment in the United States peaked in 1970 at 18 percent of the workforce. By 1991, the proportion of Americans employed as clerical workers had fallen to 16 percent. The upshot of these developments, as almost everyone knows, is that work has become increasingly white-collar and oriented to the provision of services. However, as Table 1.1 makes clear, the nature of the change is not what many discussions of the service economy imply. For instance, although the proportion of Americans classified as service workers has grown since 1950, the 3 percent increase (from 11 to 14 percent) is actually smaller than the increases shown by all other occupational clusters that grew during the same period. Thus, we submit that if the data in Table 1.1 indicate a shift toward a service economy, it does not appear to be an economy dominated by the low-wage, low-skill jobs that the government classifies as service occupations. Instead, Table 1.1 suggests that professional and technical jobs are increasing faster than all others and may become the modal form of work for the twenty-first century. The number of professional and technical jobs in the United States has grown by more than 300 percent since 1950 (see Figure 1.1). No other occupational sector has experienced nearly as great a growth rate. Even sales (248 percent) and I. The decline of unskilled and semi-skilled blue-collar labor actually began after 1940, when the proportion of Americans employed as operatives and laborers peaked at 28 percent of the labor force.

4

BARLEY AND ORR

350

311 300 248

250

200

182 161

156

ISO 97

100 50

Managerial Professional and Technical

Clerical

Services

Sales

Crafts

Total

FIGURE 1.1 Percentage growth in U.S. occupational categories, 1950-1991

managerial work (182 percent), which expanded tremendously over the last two decades, lag behind the growth of the professional and technical labor force. A quarter of all new jobs currently being created in the United States are either professional or technical in nature (Silvestri and Lucasiewicz 1991 ). As Table I.l shows, by 1991 more Americans were found in professional and technical occupations ( 17 percent) than any other occupational category tracked by the Bureau of the Census. The Bureau of Labor Statistics' most recent employment forecast indicates that professional and technical workers will represent 18 percent of the labor force by 2005 (Silvestri and Lucasiewicz 1991 ). Similar trends are to be found in employment data from Canada and Great Britain, where professional and technical workers already account for 18 percent and 20 percent of the workforce respectively (Barley 1995). Sociologists have long criticized aggregate occupational data for a variety of reasons, including the fact that government categories lump together occupations in an analytically naive, if not haphazard, way (Spenner 1980; Miller et al. 1980). This creates the possibility of either overestimating or underestimating the importance of change in any category; however, there is good reason to believe that in the case of professional and technical occupations, classification problems are likely to lead to conservative estimates of growth. Whereas most of the occupa-

INTRODUCTION

5

tions classified as professional or technical by the U.S. government also satis. fy sociological and cultural criteria for the label, numerous occupations that the government currently allocates to other categories also have strong technical components. For instance, the census classifies individuals who repair and maintain computers as crafts persons, even though they are typically called technicians in everyday life. Accountants are labeled managers, although sociologists of work and accountants themselves generally consider accounting to be a profession. Computer operators are classified as clerical and kindred workers. A sense of the extent to which census categories may undercount professional and technical workers can be gleaned from the fact that if classified differently, these three lines of work alone would raise the proportion of Americans classified as professionals and technicians by a full percentage point. 2 If problems of "misclassification" were suddenly resolved, aggregate occupational data would still provide a hazy indicator of the shifting nature of work because, by definition, such schemes are based on job titles. At best, they index changes in what people do, but they are largely insensitive to changes in how people do what they do, unless shifts in technique also lead to changes in occupational nomenclature. Aggregate occupational data, therefore, are generally blind even to systematic changes in the way work is performed. Yet, as we have noted, the Industrial Revolution merits its name not simply because people began to pursue different lines of work, but because people also began to do old lines of work in radically new ways. Although it is difficult to gauge the extent to which qualitative changes are occurring in the nature of work, mounting evidence indicates that such change may be widespread. Moreover, it appears that the change points in a consistent direction, toward what might be called, for lack of a better term, the "technization of work." By technization, we mean to characterize the emergence of work which is comparatively complex, analytic, and even abstract, because it makes use of tools that generate symbolic representations of physical phenomena and that often mediate between workers and the objects of their work. Controlling a nuclear power plant though an array of computer terminals is one example of such work; manipulating and studying the properties of cells using a cytometer is another. In some cases, technization describes the growth of new occupations. The work of sonographers and CT technicians, for example, arose de novo during the 1970s as hospitals added ultrasound and CT scanners to their arsenal.of medical imaging devices. In other instances, technization proceeds by transforming existing lines of work. The second process is usually less visible than the first because it 2. Computer and electronics repair employed 0.1 percent of all working Americans in 1991. Another 0.1 percent of the population were employed as computer operators. Accountants represented 0.8 percent of the workforce (Silvestri and Lucasiewicz 1991).

6

BARLEY AND ORR

does not change what an occupation is called and because practitioners appear to outsiders to be doing what they have always done, even though they now do it in dramatically different ways. The nature, implications, and potential scope of the second form of technization are most easily illustrated by examples drawn from lines of work that few consider technical: farming and the operation of continuous process plants. A growing number of dairy farmers are beginning to rely on computer systems to manage their herds (Price 1993). As well as using computers to keep track of a farm's finances, dairy farmers can now monitor their cattle with electronic ear tags that identify individual cows. The tags work in concert with a coordinated system of sensors to generate data on how much milk each cow provides per milking, the rate of flow, how long each milking lasts, and the amount of feed a cow consumes over the course of a day. The data is used to make decisions about the cow's diet. When a cow's milk production falls, the tag issues a warning so the farmer may summon a veterinarian to examine the animal's health. In still other applications, sensors in fields send data on soil moisture and atmospheric humidity to a computer that combines these data with data from recent weather forecasts to control irrigation. Such systems are reputed to save both water and energy while producing healthier crops and more abundant yields (Smith 1984 ). Operators in continuous process plants used to rely on sight, sound, smell, taste, and feel in addition to thermometers and pressure gauges to monitor production runs. As Shoshana Zuboff ( 1988) described in her influential study of pulp paper mills, operators now sit in elevated, air-conditioned control rooms where they monitor and intervene in production using an array of workstations that display and process data collected from hundreds of sensors spread across the factory floor. As a result, the operator's job has become increasingly analytical. Similar control technologies are rapidly altering production in other continuous process plants whose products range from peanut butter to petrochemicals.

ENGINES OF CHANGE

The Growth and Commercialization of Scientific Knowledge Several interwoven dynamics appear to have occasioned both the technization of work and the expansion of the professional and technical labor force. Of paramount importance has been the harnessing of science for commercial advantage. Over the last century, industry has come to view science as a virtually limitless source of ideas for new products, new production processes, and even new industries. The commercialization of chemistry and physics during the last two

INTRODUCTION

7

decades of the nineteenth and the early decades of the twentieth century gave rise to the industries on which most Western economies now pivot: aerospace, energy, pharmaceuticals, petrochemicals, and electronics. Advances in the life sciences, especially in immunology, microbiology, biophysics, and biochemistry, made possible the expansion of the health care industry after World War II. More recently, molecular biology and computer science have created opportunities for entirely new industries and have revolutionized others. The growth and commercialization of science has shifted the balance of employment toward technical work in three ways. First, the escalating demand for scientific knowledge, both basic and applied, has translated directly into employment opportunities for scientists, engineers, and other technical professionals. As the chemical and biotechnology industries grew, for example, so did their respective demands for chemists and molecular biologists. The demand for scientific and technical talent has done more than foster employment for members of existing disciplines, however; it has also triggered a proliferation of new fields and occupations. As a technical discipline grows, it becomes increasingly difficult for individuals to master the breadth of knowledge necessary to remain a generalist. Generalists are also less well prepared than specialists to provide "state-of-the-art" services. Consequently, most sciences and professions divide themselves into ever narrower subfields as their knowledge base grows. All else being equal, specialization should increase the number of employed professionals by requiring collaboration. Few specialists can execute alone tasks that require both breadth and depth of experience. Thus, as specialization proceeds, the number of experts necessary to accomplish a complex task, such as rendering a medical diagnosis, burgeons. The growth of scientific knowledge spawned tech. nical work by yet a third path. As fields grow, scientists and professionals tend to allocate more routine duties to somewhat less well trained individuals. As Peter Whalley and Stephen Barley note in Chapter 1, numerous technicians' occupations are rooted in precisely such a "hiving-off" process. Thus, scientific specialization encourages the proliferation of secondary support occupations which are themselves technical.

Technological Change Another critical factor in the shift to a technical workforce has been technological change. Throughout history, technologies have spawned new occupations: the wheelwright, the blacksmith, the machinist, the automobile mechanic, and the airline pilot are illustrations. In the past, technologies created occupations across the entire division of labor. Although modem technologies have also produced occupations in all strata, those with high technical content appear to have become more common (Adler 1992). The advent of computer technology

8

BARLEY AND ORR

and the subsequent shift from mechanical to microelectronic devices are largely responsible for the difference. In 1950, for instance, few people worked with computers and most who did were mathematicians (Pettigrew 1973). By the 1970s, computers had given birth to such well-known occupations as programmer, systems analyst, operations researcher, computer operator, and computer repair technician. These occupations continue to be among the fastest growing. In the United States alone they are anticipated to provide employment for 2.3 million people by the tum of the century, or 1.6 percent of the American labor force. 3 The explosion of occupations directly related to the computer, however, is only the most visible sign that technology may now favor a technical labor force. Numerous technical occupations have been created over the last four decades by technologies other than the computer; nuclear technicians, nuclear medical technicians, broadcast engineers, ~nd materials scientists are examples. Moreover, computers have altered the contours of many traditional jobs. To grasp how computers have accelerated the technization of existing work, one must distinguish between substitutional and infrastructural technological change (Barley 1991). Most technological change is perceived to be a matter of substitution, replacing an earlier technology with a more efficient or effective successor. However, the simplicity of substitution is usually illusory. Work with a new technology will rarely be done in the same way as it was with its predecessor, and so the changes will necessarily be more complex than is suggested by substitution. Nevertheless, historically, substitute technologies have usually been adopted on the grounds that they make some parts of work easier to perform. Some have also generated considerable profits by reducing labor costs and allowing economies of scale. By comparison, infrastructural change is rare. Infrastructural technologies are the relatively small set of technologies that form the bedrock of a society's system of production during a particular historical era. Until recently, the economies of the industrial nations revolved around electrical power, the electric motor, the internal combustion engine, and the telephone (Coombs 1984). The diffusion of these technologies occasioned the Second Industrial Revolution. New technologies, even those that alter the infrastructure of production, are initially perceived as substitutes for existing technologies, and microelectronics have been no exception. Early computers were adopted by organizations to streamline personnel, accounting, and other paper processing operations. Similar motives have underwritten the spread of most other digital technologies. Digital sensors, robots, production scheduling systems, and computer-integrated 3. Estimates are based on data from Silvestri and Lucasiewicz's (1991) estimates for a moderate growth scenario.

INTRODUCTION

9

machine tools, for instance, purport to make manufacturing processes more efficient by replacing slower mechanical devices, paper-based inventory systems, and semi-skilled labor. Office automation enabled firms to reduce the number of clerical personnel they employ, reputedly by allowing offices to increase the volume of data they process per unit time. Because digital technologies are often developed, bought, and sold explicitly for their substitutional promise, early sociological research on computers and other microelectronic technologies focused almost entirely on reductions in force and the deskilling of work (Zimbalist 1979; Crompton and Jones 1984; Noble 1984). Yet, even in their purely substitutional role, microelectronic technologies have furthered the shift to an increasingly technical labor force. In addition to requiring new skills of their users, these technologies require a cadre of skilled technicians capable of programming, diagnosing, and repairing devices. Barbara Baran (1987), Thomas Diprete ( 1988), and Paul Attewell ( 1987) have shown that the movement of computers into the insurance and banking industries led to a slight increase in average levels of skill primarily because the shift to computational technologies required firms to hire programmers and technicians even as they reduced the number of clerical employees. Harry Braverman (1974), who first articulated the deskilling thesis, also recognized that firms would require a cadre of programmers and technicians if they were to use numerical control successfully to deskill or replace machinists. Only recently, however, have students of technology begun to realize that microelectronics' greatest influence on the social organization of work may arise from its infrastructural rather than its substitutional potential (Adler 1992, Barley 1991 ). It is as an infrastructural technology that microelectronics becomes implicated in the technization of existing lines of work. Even in cases such as clerical work where the deployment of microelectronic technology has been perceived as pure substitution, the need to cope with operating systems, servers, networks, and applications has clearly wrought a fundamental change in daily tasks. The link between a microelectronic infrastructure and the technization of work rests on the concept of the digitization of information. Digitization has occurred in two distinct waves. The first centered on the translation of symbols into digital impulses that could be stored, manipulated, transferred, and decoded in electronic form. It is this type of digitization that underlies the most familiar uses of computational technology: word processing, numerical calculation, database management, and electronic mail. The computing infrastructure created by the technologies associated with the first wave of digitization made it possible to traffic in large datasets. The ability to amass and analyze huge databases, in tum, gave cultural credence to the notion of information as an economic good. Data were equated with information, a commodity that could be created, bought, and sold. The commodification of data as information accounts for the growth of a

10

BARLEY AND ORR

number of industries and professional occupations that are neither directly dependent on the expansion of scientific knowledge nor directly focused on the maintenance of the infrastructure itself. Examples include the financial services industry, commercial database services, and a variety of consulting services. The second and more recent wave of digitization has brought sophisticated sensors that transform physical signals into data and effector technologies that use digitized data to control machines and manipulate objects. Integrated sensing and manipulating technologies have been critical to the rise of computerintegrated manufacturing and the advent of new medical imaging devices. In the long run, the second wave of digitization may prove to be more critical for the restructuring of work than the first. The ability to convert physical signals into data that can be used to manipulate objects opens the possibility for radically new modes and relations of production. Not only do such technologies promise to reduce even more dramatically the number of humans directly involved in manufacturing, but they are likely to engender forms of work in which workers are increasingly distanced by digital interfaces from direct contact with the physical systems over which they have responsibility. Here, then, lies the essence of the technization of work: the increasing ability to intervene in the world of objects through symbols. The work of pulp paper mill operators described by Zuboff exemplifies such technization. Whereas the mode of production associated with the Industrial Revolution revolved around the production and manipulation of things, the mode of production associated with the shift to a professional and technical labor force is likely to revolve around the creation and manipulation of symbols in order to work on things. 4 WHAT ISTECHNICALWORK?

Far more difficult than pointing to the expansion of technical work is the task of defining what technical work is and how it differs from other types of work. One common definition draws credibility from the claim that technological 4. Although the growing economic importance of scientific knowledge and infrastructural technological change are the primary reasons for the burgeoning of technical work, it is important to recognize that both factors operate in concert with other social trends. As a case in point, consider the recent phenomenon euphemistically known as "downsizing." Although downsizing has not directly created professional and technical jobs, it has enhanced their prominence by disproportionately targeting blue-collar and mid-level managerial jobs for elimination. As a matter of accounting, the relative prominence of professional and technical work should increase as other occupational categories shrink. Although some commentators have argued that widespread downsizing has become possible largely because of computer technology, such accounts are suspect because computers have not yet significantly automated the work of middle managers. At best, downsizing appears to be a fashion whose unintended consequence has been to increase the complexity of remaining jobs and to enhance the prominence of the technical and professional workforce, which has so far largely escaped wholesale reductions.

INTRODUCTION

11

change is responsible for the emergence of the technical labor force. From this perspective, technical work entails either working "on" a complex technology, as in the case of repair, or working "through" or "with" a complex technology to achieve some other end. This definition is attractive because it enables one to count as technical not only the work of technicians who maintain equipment, but also work that has been technized. One drawback is that the definition provides no clear criteria for distinguishing between traditional blue-collar workers who simply operate complex machinery and those whose function has been fundamentally transformed. The principal weakness of the definition is that it does not address what is actually done at work. Another way of distinguishing technical work from other lines of work is focusing on how members of different occupations know what they know. From this perspective, technical work would be distinguished by the type of knowledge it requires. Perhaps the most common definition of technical work employs this approach: Technical work is work that requires one to understand and utilize an abstract body of knowledge. Those who adopt such an approach usually claim that technical work requires familiarity with mathematics or science, an appreciation of abstract principles that underwrite practice, or the mastery of a way of knowing that is somehow more "objective" than other ways of knowing (Barlow 1967; Hull1986). Analysts who adopt such a definition tend to associate technical work with analytic tasks and to focus on formal education as a source of training. The advantage of such a definition is that it allows one to treat the work of most professions as technical work. The disadvantage is that the focus on formal knowledge, rather than practice, excludes lines of work whose practitioners rely almost exclusively on contextual knowledge, but who are considered by almost everyone to be technical workers, for example, photocopier technicians (Orr 1996), microcomputer support technicians (Zabusky, Chapter 6), and almost anyone else whose job centers on the repair of complex technologies. More important, the approach devalues the fact that most studies of technical work have repeatedly shown that technical acumen, even among professionals, draws heavily on contextual and even tacit knowledge (Collins 1974; Knorr-Cetina 1981; Lynch 1985). A third approach to defining technical work begins with the observation that all work has a technical component. In everyday speech, we often use the term "technical" to refer to skilled practice, the use of technique to accomplish work. This everyday meaning is accurate, but for our purposes it is too inclusive. In some sense all work qualifies, and under this interpretation technical work differs from other forms of work only in degree, not in kind. Furthermore, this meaning includes only practice and does not recognize the abstract or formal knowledge which typically seems to be part of technical work. A critical aspect of our everyday understanding of technical work is that it is

12

BARLEY AND ORR

somehow qualitatively different from other types of work. Technical work sits at the intersection of craft and science, combining attributes of each that are normally thought to be incompatible. It is a cultural anomaly in which mental and manual skills coexist inseparably, if not always comfortably. For this reason, none of the previous definitions are adequate. Therefore, we submit that it is more productive to identify technical work by a loose constellation of attributes, several of which are not normally juxtaposed in traditional frameworks for classifying work. Although individual instances of technical work may fail to evince all characteristics, all cases will possess the majority. Four traits comprise the constellation we propose: (a) the centrality of complex technology to the work, (b) the importance of contextual knowledge and skill, (c) the importance of theories or abstract representations of phenomena, and (d) the existence of a community of practice that serves as a distributed repository for knowledge of relevance to practitioners. This understanding of technical work offers several advantages over the previous approaches. It enables the analyst to count as technical the entire array of occupations that fit common language definitions of technical work. Specifically, it allows us to treat most professionals and technicians as technical workers, as well as all those whose work has been technized, while excluding most managers as well as the incumbents of traditional clerical and blue-collar jobs that are better understood within an industrial framework. More important, it highlights the fact that technical work blurs the traditional boundaries between professions and crafts, and between blue- and white-collar jobs. Categories of technical work are anomalous in this combination of skill and abstract knowledge and often seem to fall into the gaps between better-known categories. Veterinary surgeons, for example, are universally acknowledged as professionals, but they work with animals and depend on manual skills. Their work, skillful cutting, is craft, but their profession is medicine. Emergency medical technicians work in the gap between the medical profession and a service occupation. They are connected to the first by a radio, their skills, and a working grasp of medicine; to the second, by their role as ambulance drivers. Engineers work between production workers and managers, on the one hand, and between craft and science on the other. Technicians, perhaps, best exemplify the anomalous status of technical work. They work as craftspeople informed by abstract knowledge, and their standing in society reflects the interstitial location of their work, as does the very etymology of the term "technician."

Technicians The word "technician" appears to have entered the English language from French during the first third of the nineteenth century as a synonym for the then more commonly used "technicist." Although derived from Greek and Latin roots

INTRODUCTION

13

that denote craftwork (Partridge 1966), "technician" initially indicated skill but had no occupational connotations and no link to science or technology. It was a term applied to artists who had mastered the technicalities of their medium. To be a technician in the nineteenth century, according to the Oxford English Dictionary, was to be "skilled in the technique or mechanical part of an art, as music or painting." Although being called a technician acknowledged skill, the term was not wholly flattering. The frequent juxtaposition of"technician" with the adjective "mere" implied a competent practitioner who nevertheless lacked artistic gifts. Real artists were born; technicians were schooled. From the beginning, technicians appear to have had ambiguous status: to be called a technician was to be simultaneously respected and denigrated. By the early 1900s, "technician" had acquired occupational overtones. Most dictionaries at the turn of the century defined technicians as individuals versed in the technical aspects of any subject, including the "practical arts," a nineteenth-century term for craft with an industrial flavor. The modern sense of "technician"-a person whose work revolves around instruments and who may require specialized training in a science or technology-did not appear until after World War II. The definition of the term in the 1947 edition of Funk and Wagnail's is particularly enlightening: ( 1) One skilled in the handling of instruments or in the performance of tasks requiring specialized training. (2) A rating in the armed services including those qualified for technical work; also, one having such a rating. The second entry indicates that the military may have been the first organization to use the term "technician" to denote a specific type of occupation. The military usage apparently diffused into the larger culture after World War II. Other dictionaries from the 1940s and 1950s support such an inference (e.g., Webster's New World Dictionary 1959). By the 1960s, references to a Military Occupational Specialty had disappeared from dictionaries, although the number of servicemen and servicewomen called technicians had increased. The fact that those who compiled dictionaries no longer felt obliged to distinguish between common and military usage suggests that the military's sense of the term had become colloquial. Thus, the Oxford English Dictionary's most recent (1989) definition of technician simply reads: A person qualified in the practical application of one of the sciences or mechanical arts; now esp., a person whose job it is to carry out practical work in a laboratory or to give assistance with technical equipment. The trajectory by which "technician" became the name for an occupational category suggests why technicians' work may be emblematic not only of the socioeconomic factors that have led to a more technical labor force, but also of the

14

BARLEY AND ORR

cultural dilemmas posed by such a labor force. Technicians' work is simultaneously associated with science and craft, which have historically stood on opposite sides of the divide between mental and manual labor. The advent of technicians' work accompanied the rise of electronic technologies, the growing economic importance of science, and, by implication, the decline of the traditional industrial regime. Perhaps most intriguing, technicians seem to have always occupied a culturally ambivalent station: to be a technician is to be highly skilled and yet to be called a technician is to be subtly maligned. We propose that this difficulty arises, at least in part, because technicians mediate between technology and society. Technicians can be said to stand between technology and society in two senses, the first cultural, the second structural. They are culturally situated between technology and society because technological change has shoved them toward the economic heart of a society not quite ready to leave behind the categories of industrialism, its distribution of power, and its presumed distribution of knowledge. Technicians stand somewhere between our future and our past, confronting us with contradictions that confuse and, at times, threaten us. As Stacia Zabusky notes in Chapter 6, technicians are expected to be servants as well as experts. They are expected to ensure that the "system" runs and to rescue us from the complexities and "normal accidents" (Perrow 1984) of the technologies we create but no longer understand. Yet, at all other times, technicians are supposed to be invisible, perhaps to support the pretense that nothing has changed and the world is still as it was. Technicians stand between technology and society in a structural sense as well, and it is here that their power lurks. They link us to technologies that are nearly transparent when they work and troublesomely opaque when they do not. They enable us to get on with our lives without knowing too much about the machinery that runs in the background but makes our lives, as we know them, possible. In this sense, then, technicians separate us from the technology on which our society is based. Barley ( 1993) has argued that differences in the way this mediation is structured underwrite a typology oftechnicians' work. Some technicians, such as science technicians, engineering technicians, sonographers, and emergency medical technicians, work as "buffers": They simultaneously link professionals to, and shield them from, aspects of the material world about which those professionals are presumably experts. In one way or another, such technicians reduce material phenomena to information that becomes grist for the mill of another occupation. Other technicians serve as "brokers": they link nontechnical communities that use technologies to the technical communities that produce them. Often brokers adapt the technologies to the needs of the user community and ensure that the system remains operational. Examples of brokers include computer technicians, photocopier technicians, and automobile technicians. Bro-

INTRODUCTION

15

kers and buffers occupy a critical role precisely because they stand between the technology in which they are expert and the segment of society that the technology serves. As Michel Crozier (1963) observed with respect to mechanics, such a position often brings latent power incommensurate with one's status. The technicians' power rests on the fact that removing them from a production system would quickly lead to the system's collapse. For this reason, technicians are paradigmatic of all technical workers. In its various forms, technical work has become socially essential precisely because it allows all other types of work to proceed. It has become the fulcrum of the emerging economy, even though technicians remain the most neglected members of the workforce. We believe that the time has come to examine technical work on its own terms and to ask whether it challenges modes and relations of production that have gone before, and, if so, how. Because the cluster of attributes that define technical work depend on the doing of the work, we think the most appropriate strategy is to study its practice. This collection of papers represents our undertaking of that effort.

DISCONTINUITIES AND DISJUNCTURES: FROMTHEORYTO PRACTICETO POLICY

The ten chapters that follow were written explicitly to take the reader on a three-legged intellectual journey. The papers in Part I comprise the first leg of the journey, exploring the position of technical workers in modem society. They highlight the cultural ambiguities that surround technical work and point to the inadequacy of existing theories for making sense of the technical worker's position in the division of labor. Peter Whalley and Stephen Barley detail the anomalous position of the technical workforce by showing that engineers and technicians violate long-standing distinctions on which our notions of work are based. In particular, Whalley and Barley argue that technical work poses a cultural problem because it bridges the chasm between mental and manual work and belies the assumption that workmust be structured according to either organizational or occupational principles. These anomalies upset the established order, causing problems of control for organizations and problems of verisimilitude for social theorists. The papers by Jeffrey Keefe and Denise Potosky and Sean Creighton and Randy Hodson further substantiate the notion that technical work is socially anomalous. Together the two papers show that technicians do not quite resemble the members of any other occupational group or, for that matter, behave in the ways social theories say they should. Focusing on aggregate occupational data as well as a case study of technicians in a pharmaceutical company, Keefe and

16

BARLEY AND ORR

Potosky demonstrate that technicians are in some ways similar to professionals, in other ways similar to craftspeople, and in yet other ways similar to factory operatives. Most intriguing is their finding that technicians are well-educated, relatively well-paid, and among the most highly skilled employees, yet they express more job dissatisfaction than any other occupational group, including semiskilled and unskilled laborers. Using survey data, Creighton and Hodson take a different theoretical stance but reach conclusions that parallel Keefe and Potosky's. Although technical workers resemble professionals in that they have high degrees of skill and enjoy reasonable prestige, their attitudes toward management are more like those of blue-collar workers. On the basis of these patterns, Creighton and Hodson suggest that technical workers be considered a "distinct category," not well described by existing sociological theory. The papers in Part I explain why we have little choice but to study what technical workers do, if we wish to understand the social organization of technical work. In Part II the reader journeys to the level of detail. Each paper in Part II focuses a trained ethnographic eye on the work and culture of a single technical occupation. Trevor Pinch, H. M. Collins, and Larry Carbone analyze how veterinary surgeons judge the difficulty of a procedure, thus arriving at a contextualized understanding of skill. Brian Pentland describes the problem-solving strategies employed by technicians who staff software support hotlines. Stacia Zabusky investigates the difficulties microcomputer support technicians encounter as they attempt to construct and maintain a technical identity in the midst of an organization overwhelmingly populated by nontechnical users. Finally, Bonalyn Nelsen describes the situated moral understandings that guide emergency medical technicians through the uncertainties and risks they encounter in the course of their practice. The vistas in Part II are less grand than those in Part I, but the resolution is much sharper. The reader will learn that the anomalies that surface in aggregate analyses of technical work in Part I are paralleled by, if not rooted in, disjunctures and ironies that pervade day-to-day life in technical settings. As we have already noted, no claim about technical work is more prevalent than the assertion that effective technical practice relies heavily on theoretical or abstract knowledge. Such claims are often used to justify requiring technicians to obtain degrees from postsecondary educational institutions in a relevant discipline. Pinch, Collins, and Carbone show that even in surgery, critical skills and judgments are contextually situated and formal knowledge is, at best, a small portion of what enables a surgeon to confront the ambiguities of practice successfully. Pentland further explores the disjuncture between formal and practical knowledge by explaining why support specialists have no choice but to approach software, the epitome of a closed rational system, as if it were contingent and poorly behaved.

INTRODUCTION

17

Zabusky examines why technicians find it so difficult to gamer respect for their skills and knowledge. She argues that microcomputer support specialists (and by implication other technicians who serve as brokers) inhabit a situation that breeds ambivalent identities. Microcomputer support specialists are simultaneously insiders and outsiders in the organizations in which they work. Their work places them in the ambiguous situation of being servants as well as experts. The pernicious dynamics of trust and distrust that surround such identities help us understand why technicians might appear in aggregate to be a cross between professionals and disgruntled employees. Yet, Zabusky hints at another dynamic that is largely obscured by aggregate analyses and that complicates any easy parallel between technicians and unskilled labor: the fact that technicians subscribe to a moral code that makes them incredibly conscientious, responsible, and even loyal employees. Nelsen's paper on EMTs complements Zabusky's analysis by explicitly examining how situated occupational ethics influence technical practice and the services that technicians provide. Nelsen shows that a shared moral code with clearly articulated notions of responsibility enables EMTs to navigate the disjunctures between autonomy and constraint and between expertise and servitude that characterize technicians' work. Moral understandings transform these disjunctures into situational dilemmas with a game-like structure of risk. The papers in Part III pivot from examinations of practice to policy, demonstrating that close studies of technical work can lead to pragmatic as well as theoretical destinations. Mario Scarselletta's analysis of the nature of skill among medical technologists and technicians indicates how a situated appreciation of the disjuncture between formal and contextual knowledge may make for better regulatory and educational policy. Specifically, Scarselletta demonstrates why regulations designed to reduce laboratory errors by requiring more education of technicians and technologists are likely to be ineffective: they are predicated on a fundamental misunderstanding of the primary sources of error in laboratory work and on a misconception of what laboratory technicians actually do. Louis L. Bucciarelli and Sarah Kuhn develop a similar theme with respect to the disjuncture between engineering practice and education. They argue that the tendency to portray engineering as an applied science, when engineering is also a social process, has allowed engineering schools to become increasingly disconnected from the exigencies that confront young graduates. Leslie Perlow and Lotte Bailyn show that the disjuncture between the practice and ideology of engineering also adversely affects firms and the engineers they employ. Because "real engineering" is usually portrayed as an individualistic accomplishment focused on the world of objects, the contributions of engineers who facilitate the social dynamics that make engineering possible often go unrewarded. As a result, engineers who enjoy and are accomplished at the social aspects of engi-

18

BARLEY AND ORR

neering find themselves channeled into careers that are suboptimal for themselves and the companies for which they work. The papers in Part III illustrate the problems that arise when decision makers attempt to regulate work and workers without considering the realities of the job. Because technical work, in particular, is so poorly understood, policy makers routinely fall back on stereotypes or images of work drawn from other occupations in order to make sense of technical work. When regulators mistake credentials for technical skill, when educators do not understand the work for which students are preparing, and when employers do not fully understand the practices and abilities necessary for technical work to proceed, they risk setting policies whose effects are not only unintended but sometimes so skewed that they exacerbate the problems they seek to resolve. The potential cost of misunderstanding technical work extends beyond educational policies for specific occupations or the policies of particular firms. In recent years, influential and well-meaning economists and government officials have argued that the United States stands at the brink of a new era and must choose between a "high road" and a "low road" economy. The former is associated with a high skill/high wage labor force, the latter with a low skill/low wage labor force. The path to the "low road" presumably leads past increased reliance on foreign labor for manufacturing goods, further downsizings of the blue-collar and managerial labor force, and an emphasis on personal service industries. Traveling the "high road" is said to require embracing the notion that prosperity lies in developing a high-technology, information-based economy. The argument is that such an economy will demand highly skilled workers. Proponents of the "high road" scenario therefore argue for policies that will raise the skill level of the labor force, since higher incomes and standards of living should "naturally" follow. The ethnographic evidence in this volume hints that such an approach may be less successful than its advocates suggest. First, between the conception and realization of a high skill/high wage economy falls the shadow of status. A number of chapters in this volume clearly indicate that skills are not necessarily recognized. As Zabusky notes, highly skilled technicians frequently find themselves cast in the role of servants. Even more troublesome, studies of technical work find only sporadic links between high skill and high wages. The salaries of some technicians lie well below that of blue-collar labor, even though no one disputes that technicians are better educated. And, as Keefe and Potosky illustrate, wage growth among technicians' occupations has not kept pace with increasing demand. In fact, in some technicians' occupations, wages have actually fallen since the early 1980s. We suggest that the misapprehension of technical work and the potential implications of its growth is rooted in the continued use of categories and theories

INTRODUCTION

19

drawn from a now-fading industrial economy to discuss the nature of work in the emergent economy. By continuing to force new forms of work into a conceptual system appropriate for the declining blue-collar and clerical workforce, analysts reduce the odds that they will appreciate the way in which work, skills, and perhaps even economic dynamics have changed. Careful examination of practice suggests a need for new categories and a reconceptualized division of labor, at least for the industrialized nations. Specifically, we see a need to re-examine the supposition of tight links between wages, skill, and status. To the extent that emergent skills do not fit existing categories, they may not even be recognized as such, much less be rewarded. To assume that policy makers can facilitate a shift to a "high road" economy by simply facilitating the acquisition of skill and the emergence of high-technology, information-based industries seems naive. Technology, work, and skills apparently change more quickly than does our repertoire of concepts for understanding either the workplace, the labor process, or what people do for a living. In this final disjuncture looms the specter of policy failure.

PART I

Technical Work's Challenge to the Established Order

TECHNICAL WORK IN THE DIVISION OF LABOR: STALKINGTHEWILY ANOMALY Peter Whalley and Stephen R. Barley

INTRODUCTION

As in all societies, Western notions of work rest on long-standing cultural distinctions, legacies of meaning which bind us to our past. Though we may not even be fully aware of them, distinctions between management and labor, profession and craft, blue- and white-collar, employee and self-employed, middle and working class, skilled and unskilled, permeate the way we talk, write, and think about work. They have utility precisely because we take them for granted. Such cultural categories shape our analyses, frame our data collection, and guide our policies and programs. They allow us to get on with the business at hand. From time to time, however, such cultural lenses may obscure more than they reveal, particularly when social contours shift in directions that differ markedly from those of the past. Under such conditions, relying on well-worn typifications may legitimate stances or courses of action out of step with the demands of the present. In this essay we argue that the rapid expansion and increasing economic centrality of technical work marks just such a time. The difficulty of locating technical work within the boundaries of conventional notions of class has been discussed elsewhere (Smith 1987; Whalley The work on technicians reported herein was supported under the Education Research and Development Center program agreement number R117QOOOII-91, CFDA 84.117Q, as administered by the Office of Educational Research and Improvement, U.S. Department of Education. The findings and opinions expressed here do not reflect the position or policies of the Office of Educational Research and Improvement or the U.S. Department of Education. The work on engineers was supported in part by a grant from the International Division of the Ford Foundation (No. 74-05940) which likewise bears no responsibility for any of the findings or opinions expressed here.

24

WHALLEY AND BARLEY

1991a). We shall focus here on the difficulties faced in trying to comprehend technical work through two other cultural frames: ( l) the categorization of work as either mental or manual which, in tum, undergirds the difference between blue- and white-collar work as well as the difference between professions and crafts, and (2) the notion that work must be either occupationally or organizationally structured. Technical work, we argue, transcends and destabilizes both dichotomies. Although societies have devised structures to manage the cultural anomalies that result, these resolutions have never been completely stable, particularly in the case of engineering. The difficulty of resolving the strains posed by technical work may become even more pronounced as the newly emerging category of technicians continues to expand.

CULTURAL FRAMES OF REFERENCE

Mental and Manual One of the oldest stories we tell ourselves about work is that there are two kinds of labor: work done with the hands and work done with the head. Manual work, contaminated by its actual and symbolic association with dirt, material objects, and physical labor has long been accorded low status. Even when manual labor is intensely skilled, it has still been devalued because of its reliance on oral traditions and tactile understandings rather than more formal and abstract codifications. Mental work, by contrast, is clean and relatively privileged, largely because it involves working with symbolic representations of a "virtual" world abstracted and distanced from nature. When mental work is considered work at ali-in preindustrial times people often viewed the manipulation of symbols as an accomplishment rather than a form of labor-it is seen as "civilized." Especially during periods when Western societies restricted access to the tools of symbolic manipulation, the ability to read, write, and do mathematics was both a source and a symbol of social status. If manual labor has historically been the work of inferiors-slaves, peasants, and, more respectably, skilled craftsmen-mental labor has been the privilege of the elite: of gentlemen, and sometimes ladies, of aristocrats, officers, scholars, professionals and priests. It was disdain for manual work that condemned poor eighteenth-century French aristocrats to near starvation lest they should soil their hands with the plow. The same disdain encouraged nineteenth century English entrepreneurs to launder their wealth and enter into the leisured lifestyle of the country gentry (Wiener 1981). In the mid-twentieth century, disdain for manual work has shaped the education of elite engineers by minimizing their exposure to practical skills to prevent such skills from lowering engineering's status in the eyes of the academic and professional establishments (Calvert 1967; Ferguson

TECHNICAL WORK IN THE DIVISION OF LABOR

25

1992; Bucciarelli and Kuhn, Chapter 9). It has also played an important role in shaping the social and political allegiances of the clerical labor force (Kocka 1980; Mills 1956) and in fostering the sense of cultural inferiority felt by many blue-collar workers whose education provided little access to cultural resources (Sennett and Cobb 1972). The superiority of mental or, perhaps better, "symbolic" work has never been simply a matter of prestige. From the beginning, writing and mathematics have been tools of social control. Writing made possible the development of the legal codes that allowed the rise of the early empires. The development of mathematics allowed the collection of levies and the keeping of accounts. These symbolic skills enabled the elite classes to "act at a distance" (Latour 1987) because they permitted abstraction from the specifics of the immediate context. Limiting access to symbolic skills further secured the power of the elite, as the medieval Catholic church explicitly acknowledged in its resistance to translating the Bible into the vernacular. Those who can use drawings, plans and other symbolic tools for controlling the manual workforce have often found places near the seats of power in industrial corporations.

Occupational and Organizational Images of Organizing A second cultural distinction that frames our understanding of work reflects the institutional context in which work is organized. Broadly speaking, we expect lines of work to be structured either occupationally or organizationally. The first term inscribes a world of crafts and professions; the second, a world of bureaucracies and firms. The former conjures up images of a relatively autonomous, collegial community of skilled practitioners; the latter, images of a hierarchy of authority and jobs of indeterminate skill linked together by wellstructured career ladders. In an occupationally ordered world, knowledge and skills are assumed to be specific to particular domains and too complex for nesting in a single hierarchy. As a result, authority and expertise are treated as the property of distinct occupational groups (Freidson 1973a). Because individuals, rather than positions, serve as vessels of expertise (Abbott 1991), knowledge is transmitted through extended apprenticeships or formal schooling in specialized curricula. Coordination occurs not through a chain of command, but through the collaboration of members of different occupations working jointly, as in the case of the modem building trades (Stinchcombe 1959). By contrast, in an organizational division of labor, authority and expertise are arranged hierarchically. People in higher echelons not only have power over those below but, at least in an ideal bureaucracy, they legitimately exercise authority only to the degree that their knowledge is seen to encompass and exceed

26

WHALLEY AND BARLEY

that of their subordinates (Weber 1968 [ 1922]). Organizations, rather than members of an occupation, become the primary means for preserving expertise because knowledge is encoded in rules, roles and procedures which are invested in positions, rather than people. Because such knowledge is often highly specific and valuable to an organization, internal labor markets evolve to define careers and ensure loyalty (Doeringer and Piore 1971). This means of organizing work epitomized the Prussian civil service on which Max Weber modeled his theory of bureaucracy, and informed Frederick Taylor's attempts to manage the modem factory "scientifically." It is also the model of work that once made "I work for IBM," a more acceptable answer to the question, "What do you do?" than theresponse, "I am an engineer" (Whalley 1987). In the preindustrial world most work was organized occupationally. Even agricultural expertise was divided according the crops that farmers grew (Applebaum 1992). The occupational system was best developed, however, among the skilled crafts whose domains were secured by guilds or guild-like collectives (Sonenscher 1987). The occupational communities that the guilds regulated were often marked by considerable specialization and an extensive apprenticeship system. Although the division of labor in preindustrial society had vertical overtones, these typically reflected either the relationship between master and apprentice or the workings of an omnipresent class structure, as in the case of serfs and lords. The organizational structuring of work was, by comparison, weakly developed. Only in the Roman Catholic church and certain European militaries was work structured along organizational lines. The modem world, by contrast, is a world of organizations, state bureaucracies, and multinational corporations. Indeed, until quite recently, most social theorists viewed the economic and political development of the twentieth century largely as a story of the ever-increasing dominance of large organizations (Chandler 1977; Schumpeter 1942; Williamson 1975). Organizations and occupations, however, are more than alternative ways of dividing labor. The contrast between them also shapes our images of work. In the United States and Great Britain, for example, the ideal of professions and crafts being the natural and desirable way to organize work has continued to exert considerable influence, even as organizational forms have come to predominate. The governments of both countries continue to collect statistics on the labor force in occupational terms (Udy 1981; Whalley 1987), and members of both cultures tend to associate autonomy, freedom, and financial independence with the occupationally structured work of the crafts and professions. In Anglo-American society, few insults are worse than calling someone a bureaucrat, with the term's connotations of colorless individuals who have sacrificed their souls to an organization. Organization men are frequently portrayed as people who are closely monitored and trapped by routines, who carry out orders without question, and

TECHNICAL WORK IN THE DIVISION OF LABOR

27

wait only for retirement. 1 In continental Europe and Japan, by contrast, the image of the civil servant, the quintessential bureaucrat, has proven far more attractive. There, bureaucracy is associated with the noble traditions of the military, the church, and the building of the state. The French "cadre" (a social category archetypically composed of middle-level bureaucrats), enjoys the same social prestige as do professions in U.S. and British society (Boltanski 1987). These distinctions-between mental and manual, and organizationally and occupationally structured work-are cultural frames of great power. They affect the way we see, think about, and value the work we do, and have important social and practical consequences for both private and public policy. The cultural gulf between mental and manual work, for example, has led to the gradual degradation of any work that requires practical skills, and to the subsequent overvaluation of white-collar jobs. As a result, educational programs have bifurcated into college-preparatory and vocational tracks: the first prestigious but entirely symbolic in content, the latter looked down upon and underfunded because of its focus on practical skills. If an increasing number of jobs come to meld mental and manual work, however, such a division may lead to a disjuncture between the society's educational policy and the requirements of its labor market. Similarly, if work can no longer be separated into jobs that are occupationally or organizationally structured, then individuals' career expectations, firms' employment policies, and society's responsibilities for education and training could all become seriously distorted. Our thesis is that the expansion of technical work poses just such a challenge to these cultural dichotomies, both practically and analytically. We shall argue that this challenge has occurred in two waves. The first began with engineering's emergence as a distinct industrial activity in the latter part of the nineteenth century. The second arose with the appearance of technicians as an identifiable occupational category after World War II. Engineers, we argue, may ultimately have found a reasonably secure but ambivalent identity as a new kind of corporate professional-although that identity, too, looks increasingly unstable. Technicians, however, continue to pose major practical and intellectual challenges to our images of work.

THE CULTURAL CHALLENGE OF ENGINEERING

Modem engineering is the child of industrial capitalism. Although commentators sometimes point to its continuities with civil and military engineering, which have much longer histories, engineering as we presently know it emerged I. The presence of women in organizational careers is still sufficiently recent to carry the glamour of pioneering, so perhaps this cultural stereotype is less applicable to them.

28

WHALLEY AND BARLEY

alongside the corporation in a world where work was still organized occupationally. In fact, engineering played an important if ambivalent role in the genesis of industrially organized labor. Although the guild system had largely disappeared by the early nineteenth century (and much earlier in England), many of its traditions lingered at the dawn of industrialization, especially among the skilled trades and learned professions. Until late in the nineteenth century control over production largely remained the preserve of craftsmen (Nelson 1975; Clawson 1980), who passed on their knowledge and skill, which was almost exclusively tacit and oral, by training apprentices (Sturt 1923). In the early days, craftsmen were often themselves employers. Those who were not often contracted with employers for control over portions of the production process, giving them authority over the work as well as unskilled laborers and apprentices. With the onset of industrialization, an organizational order slowly began to supplant this occupational order. Firms sought to rationalize the tacit, oral, and secret skills of the crafts and to locate craftsmen within a vertical, rather than horizontal, division of labor. Semi-skilled workers, whose activities were regulated by machines and engineers, gradually replaced those craftspeople who had stood at the heart of the production system: ironworkers, tailors, seamstresses, machinists and others (Edwards 1979). 2 The process involved tearing apart older, more integrated craft techniques so that design and planning could be separated from actual manufacture. It also entailed developing new and more symbolic methods that permitted the control of manufacture from afar. This separation of planning from execution led to control over the labor process being vested in managers whose technical expertise was sometimes more theoretical than practical. It also led to the establishment of drawing offices where machines and production processes were often designed and developed with the goal of tighter regulation of the shop floor (Ferguson 1992). Such a radical reorganization of work required the creation of a technical staff knowledgeable in design and production techniques, committed to the employers' interest, and willing to assist management in controlling the industrial workforce. Technical staff began to control both the speed and the methods of production, a process carried to its logical, if sometimes irrational, conclusion by the practitioners of scientific management (Braverman 1974). Later, research and development laboratories began to institutionalize the development of new products by incorporating into the organizational division of labor the work of the inde2. A few crafts successfully retained their privileges as a labor aristocracy, in part because they were more willing to cooperate with employers. Placed in charge of the maintenance of factories but located at the margins of the production process, electricians, pipe fitters, and other tradespeople specializing in building and repairing equipment, retained traditions of apprenticeship and distinct occupational identities.

TECHNICAL WORK IN THE DIVISION OF LABOR

29

pendent inventor, who, like the craftsman, had often integrated design and practice (Reich 1985, Whalley 1991b). Thus, the emergence of engineering, and the development of the industrial corporation, with its organizational division of labor, are simply two sides of the same historical process. Although modem engineering emerged hand in hand with industrialization, a serious history of the way in which modem-day engineering's various functions came to be bundled together remains to be written. 3 Certainly the military engineers of the Renaissance provided one model for the integration of design and command. Especially in France, schools of military engineering provided an important prototype: first for the civil engineers who built the roads, canals and railroads of the nineteenth century and, later and less directly, for mechanical engineering (Weiss 1982). In the United States and Great Britain, where formal education in engineering developed later (Calvert 1967; Smith and Whalley 1996), the artisans who built the early machines of the Industrial Revolution were more important role models. These early mechanics were not members of a formally educated officer class, trained in math and drawing, but practical tradesmen who only reluctantly supplemented their empirical approach, rooted in tacit and oral knowledge, with the skills of formal drawing needed for mass production. Although the story is sometimes written as if these mechanics were simply deskilled and replaced, many were absorbed into the new technical positions. They were supplemented, especially in the United States, by the relatives of entrepreneurs who saw a professional career in industry as the modem equivalent of the older aristocratic career in the military. All that these groups had in common was an interest in designing and planning and a willingness to apply that interest to the corporate organization of production. Out of this melange, engineering grew to become one of the largest and bestestablished professions of the twentieth century. Like other professions, engineering now boasts its own formal training programs in universities and its own professional societies. Engineers are relatively well paid and enjoy a degree of job security as well as considerable social respectability. Furthermore, engineering is firmly situated on the management side of the organizational divide that has traditionally separated white-collar from blue-collar, "staff' from "works," "suits" from "working stiffs" and "exempts" from "non-exempts."4 In short, en3. Such a history would require the kind of approach recently developed in constructivist accounts of the emergence of science disciplines (Latour 1987; Porter 1977), rather than one that prejudges the way in which various tasks were bundled into present-day occupations, or assumes that engineering was a potentiality waiting to be actualized by technological and social developments. The latter, unfortunately, has been the style of many histories of the profession. 4. Engineering has been institutionalized most fully as a "corporate profession" in the United States. In Britain, the intertwining of engineers with positions that would be filled by technicians in the U.S. has left many engineers with a weaker claim to professional status. British engineers are better understood as part of a larger corporate staff which encompasses a wider range of nonmanual

30

WHALLEY AND BARLEY

gineers have become what Peter Whalley ( 1986b) has called "trusted workers." They are collectively, though not necessarily individually, loyal to their company and share, in good times at least, many-though not all--of the political interests of senior management. In fact, engineering is something of a model for a range of newer corporate professions such as accounting, marketing, and personnel management. At first glance, then, engineering seems to fit our conventional cultural categories for classifying work rather well. White-collar and even professional, engineering is so tightly linked to science--often thought of as the archetypical "mental" work-that the public often finds it difficult to separate the two. Engineers have also been so central to industrial organizations that many firms considered to be icons of the American economy-IBM and General Motors, for example-are widely perceived to be embodiments of engineering culture. Yet, if we look more closely at engineering, several discrepancies emerge that are rooted in the peculiar attributes of technical work.

Anomalies Mental or Manual. Although engineering may be firmly ensconced as a twentieth century profession and the curricula of engineering schools may center on the learning and application of scientific principles, the occupation has never fully escaped its manual legacy. Engineering has never been quite abstract or symbolic enough to be a purely "learned" profession. Unlike lawyers, scientists and clergy, engineers have always been involved in practical matters, in making things and making them work. If mathematical or scientific analysis assists the engineer, all the better. But if successful practice is shown to depend on trial and error or on local and contextual knowledge, then that too has generally been acceptable to most engineers (Ferguson 1992; Henderson 1991; Vincenti 1984 ). Engineering is technical work not simply because engineers rely on esoteric techniques and instruments, but because machines and systems are engineering's ultimate product, even if, in a siliconized world, some machines have a virtual rather than a mechanical reality. Making existing machines work, designing new ones, and doing these tasks well, remains at the heart of engineering culture (Kidder 1981; Kunda 1992).

work. Nonetheless, many of the distinctive social arrangements that characterize corporate professions in the United States are also in place in Britain. These practices serve to secure corporate loyalty and separate technical staff from manual workers (Whalley 1991a). In France, engineers have an even higher status than they do in the States. As in the U.S., the French notion of cadre excludes technicians but includes a broader range of management positions and stresses organizational authority rather than technical expertise (Boltanski 1987; Crawford 1989).

TECHNICAL WORK IN THE DIVISION OF LABOR

31

This commitment to machines and practical activity has always made it difficult for engineering to depend solely on college education for training. In all but the most advanced areas of high technology, practical experience continues to play an important role. Engineers retain a grudging admiration for the selftrained technical wizard who knows almost intuitively how to design and build machines and get them to work. Studies in a variety of industrial settings and in countries with different educational systems have consistently found that about a third of all practicing engineers are former craftspeople and technicians with no formal training in engineering (Crawford 1989; Whalley 1986b; Zussman 1985). Although this percentage may be declining as older engineers retire and as pressures for credentials increase (Keefe and Potosky, Chapter 2), much credentialling still occurs after the individual has entered the occupation. This "manual" side of engineering has always been most visible in Britain. Early in the nineteenth century the Mechanics Institutes sought to integrate practice and theory by exposing artisans to the world of equations, and universitytrained engineers to the world of manufacturing (Wrigley 1986). By the end of the century, however, the middle classes, wary ofthe status implications of working too close to machines, had distanced themselves from engineering (Ahlstrom 1982; Donnelly 1991; Wiener 1981). As a result, university-trained engineers played only a minor role in industrializing Britain. The lack of university-trained engineers, in turn, encouraged British firms to rely more heavily than elsewhere on apprenticeships and promotion from within (Smith 1987; Smith and Whalley 1996; Whalley 1986b). Indeed, employers often resisted efforts to formalize engineering education, on the grounds that it would distance engineers from industrial practice. British engineers still retain strong links to craft traditions and identify more closely with the practical aspects of their work than do engineers in other countries (Whalley and Crawford 1984). Many are even members of unions aligned in some fashion with the crafts (Smith 1987). In the U.S. status barriers were less strong than in Britain and the manual attributes of engineering were initially less stigmatizing. The children of small businessmen entered engineering in larger numbers and the owners of firms were often proud of their hands on know-how (Noble 1977). The social acceptability of mechanical acumen allowed engineers to acquire social status within a "shop culture" more typically present in the crafts (Calvert 1967). Only the rise of the professional school-located in universities but accredited by professional bodies with strong employer representation-transformed engineering's culture of practical experience into one of formal education (Layton 1971). The transformation was not without its critics. Many have argued that the search for academic respectability-the conversion of engineering from a practical art to a symbolmanipulating "science"-has weakened the engineer's skills and distanced academic engineering from the employer's needs (Ferguson 1992; Bucciarelli and

32

WHALLEY AND BARLEY

Kuhn, Chapter 9). 5 As did their British counterparts, American engineering educators attempting to promote the value of a more abstract and scientific education have run into strong resistance from both employers and many practicing engineers. The practical nature of many technical tasks has not been the only source of continued tension between the professional and craft aspects of engineering. Marxist analysts, for example, have pointed to engineers' dependent status as employees, their location in staff rather than supervisory positions, their craft-like apprenticeships, and their occasional propensity to unionize, leading some writers to portray them as members of a "working class." For reasons suggested above, the claim has been most plausible for British engineers (Smith 1987), but elements critical to the working-class argument are present to some degree in all countries. For example, upon observing the radical moment that resulted in the French student revolt of 1968, Serge Mallet ( 1975) and Andre Gorz ( 1967) pronounced technical workers part of a "new working class" whose knowledge and self-confidence were sufficiently intact for them to resist incorporation into capitalism.6 Trying to ignore distinctions between engineers and manual workers, however, has proved as difficult as treating them as purely "mental" employees. Despite their practical bent, engineers value the ways their "book learning" and mathematical abilities contribute to their work and status (Robinson and Mcllwee 1991 ). Engineers jealously guard their privileged position in the workplace. They have rarely been willing to unite with others, especially manual workers, in order to bargain collectively. Moreover, the sporadic infatuation during the 1960s with employee participation among engineers often proved to be a strategy for improving their own status in managerial affairs rather than a plea for worker solidarity (Crawford 1989, Whalley 1986a). Occupational or Organizational Work. Attempting to portray engineering in solely occupational or organizational terms is no less difficult than the Procrustean exercise of forcing it to fit traditional notions of mental and manual work. In the United States and Britain, at least, engineering is usually cast as a profession. Proponents of this view point to such occupational trappings as ere5. This is not just a practical problem for educators. The "craft" of engineering has long posed difficulties for analysts who have sought to place engineering among those professions that owe their status and power to scientific knowledge (Freidson 1973a; Larson 1977). Indeed, the very attempt to use "science" as a way of distinguishing professions and crafts runs afoul of ethnographic evidence that even the "pure sciences" have their practical and instrumental, perhaps one should say "technical," side. If even the boundaries of science are socially constructed, it is hard to seek the essence of professionalism this way (Whalley 1991a). 6. Erik Wright ( 1985), writing in the American context, also saw engineering as a "new" craft, but was less sure of its place in a unified working class. He saw engineers, like the old labor aristocracy, using their skills to "exploit" other workers and obtain more than their fair share of the surplus.

TECHNICAL WORK IN THE DIVISION OF LABOR

33

dentials, professional schools and associations, and engineering's distinct knowledge base. Historically, many engineers have indeed sought to work as technical experts unencumbered by administrative and managerial responsibilities. These engineers are often more deeply committed to their technical communities than to the organizations that employ them. There is some evidence that an occupational orientation may even be increasing among engineers, especially in hightech industry (Mcllwee and Robinson 1992). Such engineers read technical journals, get excited by technical advances, and can be recruited with offers of opportunities to "play" with the latest technology. Especially early in their careers, many engineers express disdain for the administrative component of management. There have even been historical moments when engineering's organizational status has been placed in doubt by professionalizing movements (Layton 1971; Meiksins 1989). For example, despite its seeming corporate agenda, even scientific management can be interpreted as an attempt to assert engineers' unique occupational identity and interests (Stabile 1984 ). Paying too close attention to engineering's occupational and technical culture, on the other hand, can lead one to overlook engineering's organizational character. Even design, the most visible and arguably the most professional of all engineering functions, is deeply embedded in an organizational context. Design emerged as a distinct task only with the disaggregation of craftwork. The engineering drawing was itself developed as an organizing tool (Ferguson 1992), and continues to serve that role in the structuring of modem production systems (Henderson 1991 ). Furthermore, engineers have never been restricted to design. Mass production also involved the detailed planning of production processesthe function championed by Frederick Taylor ( 1911) and other advocates of "scientific management"-which gave industrial engineering, in particular, a distinct managerial aura. In France, where the state sponsored industrialization and the French Revolution banned most craft apprenticeships, engineering clung to its preindustrial origins in the French officer corps. Their training in prestigious technical schools places French engineers at the core of the managerial class (Weiss 1982). Even in Britain, university-trained engineers are increasingly involved in the management of production (Whalley 1986a). Finally, there is much evidence that even design engineers readily show interest in management when technical and financial rewards encourage them to climb the organizational ladder (Whalley 1990). The engineer's tendency to combine occupational and organizational interests has frustrated sociologists on both sides of the Atlantic. As a generation of American sociologists discovered, attempts to treat engineering as a traditional profession founder on the engineers' lack of commitment to a set of "professional" values that might put them at odds with their employers (Perrucci 1969; Ritti 1971). Unlike other professionals whose occupations establish their identities

34

WHALLEY AND BARLEY

outside of corporate employment (scientists and physicians, for example), engineers seem to be remarkably contented employees. They show no particular desire to share their knowledge with the public at the expense of the corporation, find little wrong with the profit motive, and are willing to pursue the rewards of a managerial career even when it draws them reluctantly away from technical work. If engineers are professionals, they are committed "corporate professionals" rather than members of a traditional "free" profession. Among French sociologists it has been engineering's occupational propensities that have caused intellectual problems. French engineers have traditionally been members of an elite cadre, firmly embedded in the higher ranks of state and corporate hierarchies. What Mallet and Gorz glimpsed in the 1960s, however, was the emergence of new groups of French engineers less firmly embedded in the bureaucracy. Unlike traditional French engineers who managed production facilities, the new engineers (who often worked in R&D labs or aerospace manufacturing) served as technical experts in staff positions. For many of them, being in the position of staff felt like a loss of privilege in the formal hierarchies of French corporations. The historical moment when many French technical staff joined in the strikes and demonstrations of May 1968 was short-lived, but it forced the intellectual recognition that engineering encompassed functions not adequately characterized as "management" with the term's intimations of a close alliance with employers. If neither a "new working class" nor members of the "old'.' middle class, post-sixties structural Marxists portrayed them as a "new middle class," occupying a "contradictory class location" (Poulantzas 1975; Carchedi 1977). If engineers are partly management, they are also partly workers and, perhaps more importantly, workers who possess a distinctive kind of occupational culture. Engineers thus continue to play havoc with conventional models of work even though they are the prototypical technical workers of high industrialization. Engineering is too practical and machine-oriented to be a profession of the text, yet, too numerate and abstract to be dismissed as "mere" manual work. Engineers are too organizationally comfortable and career-oriented for their work to be fully understood in occupational terms; yet, engineers are too committed to their own technical culture to be fully assimilated into the blue-suited ranks of management. Thus, even though engineers may be corporate professionals, the underlying difficulties posed for our conventional cultural frames by the technical nature of their work have not been entirely resolved.

The Future of Engineering Ultimately, the formation of engineering as a corporate profession needs to be understood as a social construction that finessed opposing cultural images of the nature of work without fully resolving the tensions embedded in them. Even so,

TECHNICAL WORK IN THE DIVISION OF LABOR

35

construing engineering in this way exacted a heavy price from corporations. Firms only retained engineers by granting them costly pay and benefit packages, promising relative job security, and creating opportunities for engineers to ascend into management. Historically, employers had a number of reasons for adopting this strategy. First, because engineering played a key role in the struggle for control of the shop floor, employers needed the engineers' loyalty. Second, because of the explosive growth of the economy during most of the twentieth century, firms could easily create enough middle-management positions to offer engineers promotional opportunities. Finally, even though engineering was the largest of the corporate professions, in all but a few high-tech companies engineers have represented only a small part of a firm's labor costs. Questions remain, however, as to what might occur if the social contract between engineers and corporations were to unravel, or if new categories of technical workers were to arise for whom managerial careers were inappropriate or too expensive. The first question seems ever more germane given the recent downsizing and reorganization of American corporations. Over the last decade, implicit promises of job security have been abandoned, middle-management positions have been decimated and firms appear ever more willing to treat engineers as temporary technical contractors. This may well have a major, but as yet unresearched, impact on the structure and culture of engineering. The second question raises the issue of the social location of that newly emergent group of technical workers: technicians.

THE ADVENT OF TECHNICIANS' WORK

Although a handful of technicians' occupations existed at turn of the twentieth century, most lines of work thought of as technicians' work came into being over the last forty years. 7 In fact, the modern meaning of "technician"-a person whose work revolves around instruments and who requires specialized training in a science or technology-did not appear in English-language dictionaries until after World War II (Barley and Orr, Introduction). In 1950, technicians comprised 1 percent of all employed Americans (Szafran 1992). By 1990, the percentage had grown to 3.4 percent. The U.S. Department of Labor estimates that the percentage will rise to nearly 4 percent by the middle of the next decade (Silvestri and Lukasiewicz 1991 ). These figures indicate that the proportion of Americans employed as technicians has grown by 240 percent since mid-century, a rate that dwarfs the expansion of all other occupational clusters charted by 7. For example, radiological technicians were recognized as members of a distinct occupation between World War I and World War II (Larkin 1983). Shapin (1989) has shown that as early as the seventeenth century, noted scientists employed individuals in their labs who filled roles roughly analogous to those of today's science technicians.

36

WHALLEY AND BARLEY

the Bureau of Labor Statistics. By way of comparison, service and professional occupations, which rank second and third for growth over the same time period, expanded by 89 percent and 82 percent respectively. 8 To be sure, technicians still constitute one of the smallest occupational categories, but the number of technicians can hardly be considered insignificant. Technicians now outnumber farmers in the American economy, and in some industries technicians are surprisingly numerous. For instance, as of 1990, American industry employed one engineering technician for every two engineers, one science technician for every two scientists, and two health care technicians for every physician. 9 In medical settings, technicians and technologists even outnumber registered nurses-yet we know less about technicians than about the members of any other occupation. Social Origins Like engineers, technicians usually work in organizations dominated by members of other occupations. However, because "technician" is a much broader occupational classification than "engineer," technicians' work is more diverse and less easily characterized than engineering. The paths by which technicians' occupations have evolved are also more numerous. Nevertheless, most technicians' occupations appear to have arisen by one of six analytically (but not necessarily empirically) distinct paths. Hiving off of Work. Medicine, law, engineering, science and most other widely recognized professions have witnessed a tremendous expansion and specialization of knowledge over the last forty years. In response, overburdened professionals have sought to lessen their workload, in part by allocating more routine tasks to members of other occupations. Many of the technicians' occupations that have flourished in the latter half of the twentieth century originated in the hiving off of work by the established professions. 10 The phenomenon has been most visible in health care, where licensed practical nurses, medical technologists, sonographers and an expanding array of other technicians have coalesced into occupations around tasks discarded by physicians and registered nurses. The dynamic is also prevalent outside health care, where it has given birth to a plethora of occupations ranging from the well-known (paralegals, computer programmers) to the amazingly obscure (test and pay technicians; see Kurtz and Walker 1975). 8. We have calculated these growth rates using data on an occupation's share of the labor force found in Szafran (1992) and Silvestri and Lukasiewicz (1991 ). 9. These ratios were estimated using detailed occupational census data for 1990 published in Table 2 of Silvestri and Lukasiewicz (1991 ). 10. The term "hiving off' is adopted from Smith (1987). The concept, though not the term, entered the sociological literature with Hughes (1958).

TECHNICAL WORK IN THE DIVISION OF LABOR

37

De Novo Creation. Other technicians' occupations have been created de novo in the sense that their work was not previously performed by members of an existing occupation. Almost all instances of de novo creation are rooted in the invention and subsequent commercialization of a technology. Prior to the invention of photocopying, there were no copy repair technicians (Orr 1991 b, 1996). Prior to the diffusion of microcomputers, there was no need for microcomputer support specialists. The spread of television similarly stimulated a market for television repair. Although many technicians' occupations that have arisen de novo center on the maintenance of a technology, not all do. For instance, sonographers rarely repair ultrasound equipment, yet their work arose with the use of ultrasound in medical imaging (Barley 1990). Air traffic controllers, EEG technologists, and technicians who monitor the controls of nuclear power plants are further examples of occupations, spawned by new technologies, that have little role in the technology's maintenance. Occupationalization of Amateur Work. A third group of technicians owe their positions to the occupationalization of amateur work. Radiological technologists are a case in point. At the tum of the century, anyone who could afford a cathode ray tube could establish a business providing x-rays for medical diagnosis (Larkin 1983). Not only did numerous engineers and physicists set up shop as purveyors of medical images, but after World War I medics trained to operate xray equipment on the battlefields of Europe returned home to establish medical imaging practices. Gerald Larkin ( 1983) reports that in the years following World War I, it was fashionable for people in European and North American cities to frequent such shops to secure x-rays as curiosities. The occupation of the radiological technologist arose, in part, as physicians lobbied for laws that drove amateurs and free lancers from the market (Barley 1986). Emergency medical technicians are a more recent example of a technical occupation that evolved from amateur work (Metz 1981). Before the 1970s no EMTs existed. Emergency rescue services outside of large cities were almost always staffed by volunteers charged with transporting the sick and injured rather than with performing triage. At best, members of rescue squads were trained in first aid. In response to a public outcry over the increasing number of highway fatalities, during the 1970s the federal government urged the states to create a corps of trained emergency personnel modeled after the military paramedics who had proven themselves capable of dramatically reducing mortality rates in Vietnam. All states now train and certify emergency medical technicians. Although many EMTs remain volunteers, for a variety of reasons the work is rapidly shifting from volunteer to paid personnel for whom emergency medicine has become a vocation (Nelsen and Barley 1993). Upgrading of Mechanics. The upgrading of mechanics and other craftspeople constitutes a fourth process by which technical occupations have formed. Steven

38

WHALLEY AND BARLEY

Shapin (1989) has shown that scientists employed laboratory assistants as early as the seventeenth century. Until recently, however, laboratory assistants were usually mechanics who specialized in the fabrication of instruments and other experimental apparatuses. In fact, many laboratories continue to employ machinists. The role of the modem science technician emerged as lab mechanics assumed responsibility for the design and execution of experiments. Separation of the two occupations solidified in the early twentieth century when it became routine for organizations to require technicians to have an advanced degree in a relevant discipline (see Keefe and Potosky, Chapter 2). With the occupation established, a hiving off process has more recently driven the expansion of the science technicians' duties and skills. A similar phenomenon surrounds the birth of engineering technicians and draftsmen. Peter Whalley ( 1986b) notes that, even today, firms often draw engineering technicians from the ranks of machinists who have demonstrated analytic talent. Technization of Work. A fifth and more recent impetus for the formation of technicians' occupations has been the technization of work-the transformation of traditional blue- and white-collar work by microelectronic technologies (Barley 1993). Accumulating evidence indicates that complex information systems infuse nontechnical and even semi-skilled work with technical content. Shoshana Zuboff (1988) concluded from her studies of paper mills that computer-integrated production requires blue-collar operators to analyze data and make decisions based on their analyses in order to control production processes. In the past, such skills were reserved for middle managers. Analogous findings are common to most other studies of manufacturing plants that have adopted computer-integrated controls (Hirschhorn 1984; Majchrzak 1988; Kern and Schumman 1992). The central dynamic in such settings appears to be the need for operators to rely on digitally encoded symbols rather than sensory data for monitoring and managing production systems. Regrading for Social Control. Finally, issues of social control have stimulated the creation of some technicians' positions. A growing number of firms have begun to reclassify the work of factory operatives. The regrading of work in the absence of technical change entails little more than awarding operatives a new title, sometimes in the hope of avoiding unionization. The origins of Proctor and Gamble's practice of calling its factory operatives "technicians" is a case in point. 11 During the early 1970s, a period of relative labor unrest, Proctor and 11. The following information was contained in a lecture given to a group of academics (including the second author) attending a three-day workshop on Total Quality Management practices at Proctor and Gamble, June 1-3 1993. Many of Proctor and Gamble's production lines now depend on advanced microelectronic controls. As a result, technicians' jobs appear to have been transformed through the technization of work and may now warrant the name they were originally given for other purposes.

TECHNICAL WORK IN THE DIVISION OF LABOR

39

Gamble decided to build a number of new plants in right-to-work states in the South. The firm formally titled the operators who staffed these plants "technicians." Although the southern "green field" sites were equipped with highly automated technologies, management's decision to call operators "technicians" was unrelated to the requirements of their jobs. In fact, the firm did not reclassify operators in unionized plants in the north who worked with similar technologies. The firm dubbed southern workers "technicians" as part of a plan to forestall unionization since technicians are typically counted as "exempt" employees who fall outside the purview of U.S. labor law.

Diversity and Commonality in the Social Organization ofTechnicians' Work Since no collective can escape the constraints of its past, we would expect the diverse origins of technicians' work to influence how technicians are trained, the structure of their careers, the probability that they will unionize, their social identities, and so forth. For instance, all else being equal, technicians whose work has been hived off from a profession should be most inclined to emulate professional forms of organizing. In fact, most of the technicians' occupations that use credentials as barriers to entry and have their own journals, training programs, and occupational associations are found in medicine, where hiving off has been most common. In contrast, occupations created de novo by a new technology should have relatively few barriers to entry and little formal structure. Occupations that focus on the repair of new technologies are especially likely to be open to anyone with mechanical acumen. For this reason, one might expect many people who repair computers to be self-trained. When the technology that occasions the de novo creation of a technical occupation becomes sufficiently widespread, occupational identities and fledgling communities of practice may evolve (Orr 1991b). On average, however, such occupations should evince fewer trappings of a profession than occupations created by a hiving off process. Compared to other technicians, those whose work has been altered by a process of technization or who perform duties formerly performed by amateurs should have the greatest difficulty establishing a recognized occupational identity, albeit for different reasons. Lines of work that have been infused with technical content by microelectronics generally have previously existing identities and statuses. Because those who do not do the work are unlikely to appreciate how the work has changed, perceptions of the technicians are likely to be constrained by existing cultural frameworks. Zuboff ( 1988) reports precisely such a dynamic in her study of pulp paper mills where many managers were unable to admit that the operators' work had been radically transformed by computer-

40

WHALLEY AND BARLEY

integrated manufacturing. These managers were uncomfortable granting operators greater responsibility and the operators themselves were hesitant to shed well-understood blue-collar identities and attitudes. Occupations arising from amateur work may also have more trouble acquiring a skilled identity than occupations with different origins, because they face publics who are not only unaccustomed to distinguishing between amateurs and full-timers, but who may be hesitant to attribute skill to an activity formerly performed without pay. (Nelsen and Barley 1993). Consequently, conscious attempts to construct skilled identities should be most common among technicians involved in occupationalizing amateur work. Finally, one would expect workers who become technicians by regrading alone to exhibit relatively traditional blue-collar and lower white-collar identities, since a name change is all that has occurred. Among such technicians, employment relations should be dominated by bureaucratic control, and unionization should be most common. Career lines, if they exist at all, should terminate at first-level supervision. Training should be confined to on-the-job learning and should not differ significantly from that typical of niost blue-collar or clerical work. Exceptions to this pattern should occur only when employers couple the regrading of work with an attempt to redesign work systems and employment practices. Although technicians' occupations vary with respect to origin and social organization, most still share a number of critical characteristics (Barley 1992). With the exception of factory operatives whose jobs have been regraded, few technicians are directly involved in the production of goods. At most, technicians repair or monitor the technologies that undergird production systems. Since few people who have not worked as technicians know what technicians know, their work generally appears esoteric to outsiders. Hence, technicians are usually recognized as skilled individuals. But most important, individuals employed as technicians almost always work on, with, or through reputedly complex technologies and techniques to manage an interface between a larger work process and the materials on which the process depends. Technicians traffic in symbols, data and diagnoses. They assist professionals, vend information, provide consumers with sophisticated services, and maintain complex technologies. Yet, unlike the popular image of a knowledge worker whose work is entirely symbolic, technicians also remain intimately involved with the material world. Technicians work routinely with machines, human bodies, and a host of other physical systems. Their encounters with the material world are as likely to be physical as they are to be mediated by instruments, algorithms, and techniques. Thus, technicians violate the same cultural categories that engineering challenged over a century ago.

TECHNICAL WORK IN THE DIVISION OF LABOR

41

The Anomalies of Technicians' Work Mental versus Manual. Like engineering, technicians' work blurs the boundary between mental and manual work and, by extension, the dividing line between blue- and white-collar labor. Technicians often wear white collars, carry briefcases, and conduct sophisticated scientific and mathematical analyses. Yet they use tools and instruments, work with their hands, create objects, repair equipment and, from time to time, get dirty. Like those in positions of authority in organizations, technicians have considerable autonomy and are often trusted by their employers (Barley and Bechky 1994). Yet, like those in lower echelons, technical workers are often poorly paid (Franke and Sobel 1970), are accorded low status (Zabusky and Barley 1996), and may be subject to stringent controls (Orr 1991b). As in engineering, the technician's melding of mental and manual work also blurs the distinction between craft and profession. Table 1.1 describes this blurring by situating technicians' work with respect to characteristics that sociologists traditionally attribute to crafts and professions. Technicians resemble professionals in that their work is esoteric enough that few outsiders can claim to possess their skills or knowledge. It is also relatively analytic and often requires specialized education. In fact, with the exception of professionals, technicians are the best-educated of all occupational groups (Carey and Eck 1984). The average technician in the United States now possesses two years of postsecondary education. Like the professions, some technicians' occupations have developed occupational societies and journals. Yet, in other ways, technicians' work more closely resembles a craft. Apprenticeships and on-the-job training play a crucial role in the education of technicians, just as they do in the training of craftspeople. In fact, many technicians are trained solely through apprenticeship. For instance, Stacia Zabusky and Stephen Barley ( 1996) found that over half of the microcomputer technicians they studied had no formal training in computers or electronics. Like craftspeople, technicians operate equipment, create artifacts, and possess valued manual skills. Technicians also resemble craftspeople in that they are more likely to unionize than are most other nonclerical white-collar employees, a tendency that is especially strong in Europe. Organizational versus Occupational Alignment. Like engineering, technicians' work challenges the traditional dichotomy between vertically and horizontally organized labor. Most technicians' occupations originated in an organizational context and most technicians continue to be employed by firms, hospitals and other vertically structured organizations. It is impossible to estimate with existing data the number of technicians who work for organizations populated entirely

Characteristics of professions, crafts, and technical occupations Professions

Crafts

Knowledge and skills are esoteric and well guarded. Few outside the occupation have more than a trivial understanding of the content of the occupation's knowledge base. Tasks are heavily weighted toward mental and analytic work.

Basic skills and knowledge are widely held by persons outside the occupation. However, finesse is less widely distributed.

Knowledge and skills are In some instances, amateurs exist, but in general, tively rare.

Tasks are heavily weighted toward manual and sensate skills.

of formal education of training and socialization of on-the-job trainmeans of training and socialization

Most require either specialized undergraduate or graduate training. All require a college degree. Although informally important, clearly of secondary relevance.

May require a formal apprenticeship. Otherwise, formal education is irrelevant. The primary avenue by which neophytes enter the occupation.

formal occupational organization

Professional societies, licensing, accreditation boards, professional journals are nearly universal. Required to practice

Unionization is common but not universal.

Tasks involve a heavy mental analytic component, also often has a signficant or sensate component. Most require either a bachelor's degree or a specialized degree or its equivalent. Frequently reported as a of training. The primary training in some technical tions. Mixed picure; some technical pations have journals sional societies, others Common among technical tions in medicine. Low, with the exception occupations in medicine. Less common than among more common than among professions.

Attribute

knowledge possessed outside occupation

mental/analytic vermanual/sensate work

certification occupational means of entry unionize

High Low

Not required to practice Low, primarily through union control of apprenticeship programs. High

Technical Occupations

TECHNICAL WORK IN THE DIVISION OF LABOR

43

by other technicians and related occupations, who might therefore be said to work in an occupationally structured milieu. However, it is possible to estimate the number of self-employed technicians using census data. Table 1.2 displays the percentage of individuals in each of the Bureau of Labor Statistics' nine general occupational categories who reported being self-employed at the time of the 1990 census. Only 2.5 percent of the technicians in the United States report being self-employed. In fact, the data on self-employment suggest that technicians are particularly likely to work in an organizational context: with the exception of clerical workers, no other occupational group is less likely to be self-employed. At the same time, however, there is growing evidence that technicians may be less well integrated into a vertical division of labor than their employment context would otherwise suggest. A number of technicians' occupations have developed institutional forms similar to those of the established professions. For instance, radiological technologists and medical technicians have their own licensing procedures, occupational associations, journals and training programs. More important, even technicians whose occupations are less formally organized appear to orient themselves as much to their communities of practice as they do to the organizations by which they are employed. Synthesizing data on careers from a number of ethnographies of technicians' work, Zabusky and Barley ( 1996) report that few technicians desire hierarchical advancement and those who do typically cease to seek managerial duties once they have obtained supervisory responsibility for a small group of technicians. Instead, most technicians either remain in the same job for long periods of time or else move from organization to organization in search of greater technical challenge and respect for their abilities. Zabusky and Barley ( 1996) found that, like professionals and craftspeople, technicians almost always define career development in terms of knowledge, acumen, and skill. Technicians also argue that expertise can only be TABLE 1.2 Percentage of self-employed individuals in various occupational categories Occupational Cluster

Percentage Self-Employed

Agriculture, forestry, fishing, and related occupations Marketing and sales occupations Executive, administrative, and managerial occupations Precision production, craft and repair occupations Professional occupations Service occupations Operators, fabricators, and laborers Technicians and related support occupations Administrative support occupations, including clerical All occupations

39.4

Note: Data are drawn from Table 6 in Silvestri and Lukasiewicz (1991).

13.0 12.8 11.9 9.2 6.4 3.2

2.5 1.5 8.3

44

WHALLEY AND BARLEY

accurately assessed by others in the field and that managers rarely understand technical issues even though they often have the authority to make or dispute technical decisions. Accordingly, technicians as diverse as microcomputer support technicians, medical technicians, science technicians and copier repair technicians routinely portray management and hierarchy as impediments to their work and argue that organizations rarely provide them with the respect they believe they deserve. Thus, it seems that many technicians orient their identities and careers to the occupational rather than the organizational communities to which they belong; but because most technicians also articulate and subscribe to an ethos of professionalism that emphasizes responsible expertise, they rarely pit themselves against their employers (Nelsen and Barley 1994). Technicians may detach their identities from the organizations that employ them, and may even view managers and administrators as fools-but most remain deeply committed to the quality of the services they provide to whomever they consider to be their clients. For this reason, they often speak of wanting to contribute to the good of their employing organization, even as they distance themselves from the organization's system of reward and control.

Attempts at Resolving the Anomalies Although there have been few studies of how employers, educators, and unions perceive or treat technicians, existing evidence suggest that all are well aware of the problems that technicians pose for traditional conceptions of the workforce. Particularly salient has been the fact that technicians blur the boundary between mental and manual work. In the earliest investigation of employers' perceptions of technicians, James Brady eta!. (1959) reported that some firms treated engineering technicians as if they were craftspeople, while others considered them professionals. Labor practices and employment relations varied accordingly. Brady et a!. (1959) used such differences to justify their recommendation that scholars distinguish between "industrial" and "engineering" technicians. They claimed that although both types of technicians supported engineers, the former relied primarily on "manipulative skills" while the latter required "depth and understanding of engineering and scientific principles." Whereas Brady eta!. ( 1959) claimed that firms have tried to resolve the anomalies that technicians pose by merging technicians into either the blue-collar or the professional labor force, other studies indicate that some firms view technicians as "intermediate" workers. Summarizing interview data with executives of engineering firms, the authors of an early study of technicians in the Silicon Valley wrote: "Technicians have made their particular contribution as members of the professional team. They engage in work that requires some of the know ledge

TECHNICAL WORK IN THE DIVISION OF lABOR

45

and skills of both the professional and skilled craftsman.... They have some 'know why' [theory] and 'know-how' [practice] and serve as a vital connecting link between the two groups .... Science and engineering technicians are semiprofessionals ... whose work requires competence in one of the fields of science or engineering and lies between that of the skilled craftsman and the professional" (California State Department of Education 1964, 10). As the president of a small solid state physics company put it, "These science and engineering technicians, as you call them, are not new to industry, but they have created a new problem because they now occur in large enough numbers to be recognized as a group unto themselves" (California State Department of Education 1964, 13). Commentators who portray technicians as an intermediate group generally resist forcing technicians into either mental or manual occupational categories. However, they preserve the long-standing cultural image of a hierarchical relationship between mental and manual work by situating technicians between the two. The U.S. labor movement has also generally dealt with technicians as if they were either craftspeople or professionals, in part because no American labor union exists specifically for technicians. Nevertheless, troubles arise regardless of the type of union with which technicians affiliate themselves. For instance, in 1957 the Engineers and Scientists of America (a now defunct association of engineers' and scientists' unions, founded in 1952) split over the role of technicians. One faction felt that allowing technicians into a bargaining unit compromised "professional objectives." Another argued that one could make no real distinction between professionals and technicians and that technicians were necessary if bargaining units were to have adequate strength (Brady et al. 1959). Technicians have usually fared no better when they have joined industrial or craft unions, where their professional orientation and their tendency to be more accepting of the organization's perspective places them in conflict with the larger membership. For example, in recent years some local chapters of the Communications Workers of America have actively excluded technicians from their bargaining units because they are perceived as having interests that differ from those of the rank and file. 12 Educators who design curricula for the training of technicians have also wrestled with the technician's place in the division of labor in parallel terms. Melvin Barlow (1967) prefaced his discussion of how to design technical curricula by noting that technicians have more "manipulative skills" than engineers and scientists and more "technical skills" than craftspeople. Like employers who portray technicians as an "intermediate group," Barlow concludes that the mixture 12. This information was communicated during a discussion that occurred during a conference of CWA members held at the New York State School of Industrial and Labor Relations at Cornell University in October 1993.

46

WHALLEY AND BARLEY

of skills suggests "the approximate location of the technician in the hierarchy of occupations." More recently, Daniel Hull (1986) has proposed that educators adopt an almost Comptian view of the organization of technical work: "scientists" provide the basic knowledge that "engineers" use to design devices which "technicians" test and repair and "operators" run and monitor. In Hull's model, knowledge flows down a hierarchy of sophistication from scientists to operators as skills become increasingly manual. Since technicians' work lies somewhere between professional (mental) and blue-collar (manual) work, Hull advocates that technical schools design their curricula accordingly. Managers and administrators of organizations that employ technicians also appear to be aware that technicians threaten the underlying premise of bureaucratic control by introducing elements of a horizontal division of labor into avertically organized context. Zabusky and Barley (1996) found that administrators were often willing to admit that they could not easily evaluate the work of microcomputer support technicians because they lacked the necessary expertise. As a result, administrators tended to grant the technicians considerap}e operational autonomy. At the same time, however, they insisted that technicians could not be completely trusted because of their technical orientation. As one administrator put it, "They look at themselves as professionals and feel that they alone have the capacity to judge their work. But they can't really be regarded as professionals because they do not have a broader capacity for judgement." Similarly, Barley ( 1984) found that radiologists readily admitted that sonographers often knew as much as (or more than) they did about the interpretation of ultrasound images. As a result, radiologists generally treated sonographers with considerable respect. Nevertheless, the radiologists were quick to prohibit sonographers from providing referring physicians with interpretive information, lest the sonographers' expertise undercut their own position of authority in the traditional hierarchy of medical occupations. Practitioners have sought to resolve the anomalies that technicians pose for traditional images of the workforce in much the same way that an earlier generation of sociologists sought to make sense of engineering. Technicians are either allocated to existing occupational categories or else portrayed as hybrids whose work is located in the interstices between those categories. Usually, attempts are also made to force technicians into a status hierarchy, which may be either organizational or occupational in nature. Given that employers have been unable to come to grips with the unique aspects of the technician's role, it is perhaps unsurprising that until recently most sociologists who have studied technicians have offered similarly conflicted interpretations of the technician's place in the division oflabor. William Evan (1964), for instance, concluded that engineering technicians were "marginal" workers, while B. C. Roberts et al. (1972) argued that technicians were best viewed as "intermediate" workers.

TECHNICAL WORK IN THE DIVISION OF LABOR

47

TOWARD A PRACTICE-BASED MODEL OF TECHNICIANS' WORK

The difficulty with such attempts to resolve technicians' anomalous status is that they all rely on concepts developed to characterize work in a previous era. A more promising strategy would be to develop a model of the technician's role based on the work that technicians perform. Historically, most lines of work have revolved around the manipulation of physical entities, the manipulation of symbols, and the management of people. As we have noted, the everyday meanings of "blue-collar" and "white-collar" reside largely in the gulf between these two foci. Managerial work is distinguished from other white-collar work because it also involves interpersonal communication or the manipulation of people (Zuboff 1988). Like engineering, technicians' work violates the cultural segregation of the material and the symbolic. However, unlike engineers, technicians have acquired little or no responsibility for people and are much less implicated in the control strategies of the firm. Moreover, as we shall see, technicians have a different relationship with the material world than do either engineers or craftspeople, and a different relationship to symbol systems than do professionals. Studies have repeatedly shown that technicians work at the empirical interface between a world of physical objects and a world of symbolic representations. Using sophisticated technologies and techniques, technicians orchestrate the connection between the two. Technicians act as the link between a larger system of work and the materials on which the system depends. Depending on context, the materials of relevance may be hardware (Orr, 1990), software (Pentland 199la), microorganisms (Scarselletta, Chapter 8), the human body (Nelsen and Barley 1993; Barley 1990), a manufacturing process (Hirschhorn 1984), or a variety of other physical systems. Similarly, depending on context, relevant representations may consist of data, test results, images, diagnoses, or even theories. As depicted in Figure 1.1, the technician's task at the empirical interface pivots on two complementary processes: transformation and caretaking. On one hand, technicians employ technologies and techniques to transform aspects of the material world into symbolic representations which can be used for other purposes. For instance, technicians in medical settings produce images, counts, assays and other data useful for medical diagnoses (Barley 1990; Scarselletta, Chapter 8, Nelsen, Chapter 7). Technicians in nuclear power plants (Hirschhorn 1984) and other automated facilities (Zuboff 1988) create and monitor flows of information on production systems. Science technicians reduce physical phenomena to data or "inscriptions" from which scientists construct arguments, papers and grants (Latour and Woolgar 1979). Yet technicians do more than generate symbols and information. Most are also responsible for taking care of the physical entities they oversee. Technicians are charged with ensuring that machines, organisms and other physical systems re-

48

WHALLEY AND BARLEY

Material entities

Empirical interface

Representations

Transformation

Biological systems Physical systems Mechanical systems

Technologies Techniques Knowledge

Data Test results Images Diagnoses

Caretaking

FIGURE 1.1 Technicians' work at the empirical interface

main intact and in good working order. Caretaking often requires that technicians employ the very representations they create. Thus, emergency medical technicians perform medical interventions based on diagnoses made at the site of an accident (Nelsen and Barley 1993). Technicians in biology labs employ the data they generate to husband organisms and monitor technologies (Barley and Bechky 1994). Microcomputer technicians use the results of tests and probes to alter the functioning of computer systems. The dual processes of transformation and caretaking that define the core of technicians' work also generate its anomalous standing. To be effective, technicians must manage troubles at the empirical interface to ensure the smooth production of representations as well as the integrity of a physical system. To achieve this task, technicians must comprehend the principles of the technologies they employ and make use of abstract and systematic bodies of knowledge. In this respect, technicians resemble professionals. Science technicians, for instance, require knowledge of mathematics as well as knowledge of the science on which their practice is based (Barley and Bechky 1994). Emergency medical technicians, sonographers, and medical technologists require knowledge of biological systems, pharmacology, and disease processes to render diagnostic-ally useful information. Even microcomputer support technicians require an abstract understanding of how computers and software function. Without such formal knowledge, technicians cannot perform their caretaking function optimally. But because technicians manipulate physical entities to achieve practical ends, they must also possess extensive contextual knowledge of materials, technologies, and techniques. Contextual knowledge is largely particularistic, acquired

TECHNICAL WORK IN THE DIVISION OF LABOR

49

though practice and difficult to verbalize-much less codify (Pinch, Collins, and Carbone, Chapter 4 ). It resides in a practitioner's ability to find and interpret subtle visual, aural, and tactile cues where novices see no information at all. In this sense, technical work seems to resemble a craft. Craftspeople have long been valued for their ability to render skilled performances based on an intuitive feel for materials and techniques (Harper 1987; Becker 1978). Studies show that technicians pride themselves on, and are prized by others for, precisely the same sort of skill. For instance, Alberto Cambrosio and Peter Keating (1988) report that technicians in monoclonal antibody laboratories are often unable to fully articulate their techniques for producing viable hybridomas. Consequently, labs frequently cannot duplicate each other's work even when procedures are meticulously documented. The transfer of technical knowledge requires one laboratory to dispatch technicians to another, and even then, the recipient of the information may be unable to cultivate the cell line. The importance of contextual knowledge is a theme that runs though most studies of technicians' work (see Orr, 1996, Pentland 1991a, Scarselletta 1992, Collins 1974, Jordon and Lynch 1992). The integration of formal and contextual knowledge in the course of manipulating objects and symbols simultaneously is precisely what makes technicians' work culturally anomalous. In this sense, technicians' work is neither a craft nor a profession and, hence, all such comparisons obscure the technician's role. Although technicians' work does involve operations on material entities, it is not a craft because technicians do not manipulate or transform materials to produce objects. Instead, they manipulate materials to produce symbolic representations. Technicians' work is not like that of traditional professions, despite the fact that technicians work with symbolic representations, because they do not use these representations to create other representations. Instead, they use representations to manipulate objects. Nor is technicians' work a hybrid in the sense that it is like a craft or blue-collar work on one dimension and like a profession or white-collar work on another. Instead, technicians' work punctures the existing cultural bulwark; it is at once a synthesis of mental and manual, clean and dirty, whitecollar and blue-collar. Such a synthetic melding of cultural opposites has been previously approximated only by engineering, surgery, and other "manual" professions. Similarly, technicians' work, like that of engineers, is not easily subsumed by either organizational or occupational images of structuring labor. Most technicians work in organizations that routinely attempt to fold technicians into avertical division of labor via formal job classifications, graded pay scales, and related human resource practices. Yet technicians remain strongly oriented to their community of practice. More important, technical work routinely undermines bureaucratic systems of control and stipulated roles because technicians'

50

WHALLEY AND BARLEY

expertise and knowledge is a critical resource which few outsiders possess. Thus, technicians' work brings the underpinnings of an occupational division of labor directly to the core of well-established organizational divisions of labor.

CONCLUSION

What, then, are we to make of technicians, engineers, and technical work in general? Perhaps the most that can be said, at present, is that technical work creates conditions conducive to a more horizontal division of labor where expertise is balkanized into substantive domains that cannot be easily classified using only traditional images of mental and manual work. As such, technical work has posed a continuing challenge to the vertical divisions of labor that are the legacy of the Industrial Revolution. Such organizational structures ultimately triumphed over an earlier occupational division of labor by relegating manual work to the lower echelons of a hierarchy of authority and assigning symbolic and interpersonal work to the upper echelons. The feat was accomplished by separating execution from cognition and claiming that abstract knowledge provides a platform for directing the work of others. Much of engineering's professionalizing strategy has been an attempt to make itself into the organizational bearer of much of that abstract knowledge. New questions arise, however, concerning the future of technicians' work, which is developing in a very different sociocultural and economic environment from the one present during the rise of engineering. In the long run, organizations may conceivably defuse the cultural challenge posed by technicians' work by absorbing technicians into the vertical division of labor. Attempts to treat technicians as either craftspeople or corporate professionals can be viewed as precisely such a move. It is also possible that technicians' work could serve for some as an apprenticeship to more fully professionalized positions, as has historically been the case in British engineering. However, it seems unlikely that organizations will be able to finesse the challenge posed by technicians in the same way that they once dealt with the anomalies created by engineering. Engineering could be more easily linked to management because the engineer's expertise in designing systems can be used to control the work of others. In contrast, most technicians' work is not about design, offers organizations no direct competitive advantage, and rarely carries responsibility for the work or well-being of people. Firms transformed engineering into a corporate profession in part by offering engineers managerial careers, healthy salaries, and related benefits. In an era when many organizations are downsizing and flattening their hierarchies, firms are unlikely to possess the resources required to create internal labor markets and convincing loyalty packages for technicians. In fact, most technicians' jobs are

TECHNICAL WORK IN THE DIVISION OF LABOR

51

currently divorced from the vacancy chains of their organizations. Instead, many technicians build their careers by seeking increasingly more challenging problems, a strategy which often requires that they move from organization to organization. Other technicians pursue careers by abandoning their current occupation to seek training in a related field (Zabusky and Barley 1996). For example, an appreciable number of emergency medical technicians eventually become police officers or firefighters, while a smaller number become nurses or doctors. Medical technologists either leave the field altogether or else migrate to other health care occupations. Engineering's status as a corporate profession was also bolstered by its ties to the university. In contrast, most technicians' jobs require no more than an associate's degree from a community college or technical institute. Because associate's degrees have far less status than bachelor's degrees, it is unlikely that technicians will be able to use their formal education to justify the stature of corporate professionals unless society's image of the associate's degree changes. Given these conditions, three scenarios for the future of technicians' work seem plausible. One possibility is that cultural understandings of mental and manual work may prove sufficiently malleable to mesh technicians' work into a vertical system for structuring labor and allocating status. Within such a scenario, technicians would most likely become the postindustrial analogue of the bluecollar worker. The fact that technicians are currently paid less well and accorded less status than their education and contributions warrant would seem to foreshadow such a resolution. The continuing decline of semi-skilled and unskilled work may also favor such a resolution, since the demise of blue-collar work would heighten the saliency of technical work's manual and contextual components. If technicians are transformed into the postindustrial equivalent of blue-collar workers, policy makers and employers should expect the current difficulty of convincing youth to invest in the education necessary for technical jobs to become even greater. It is unlikely, however, that technicians' work could be structured in precisely the same way that semi-skilled and unskilled labor are structured in today's economy. Not only do the attitudes and orientations of technicians differ in important ways from those of blue-collar workers (Creighton and Hodson, Chapter 3), but the symbolic products of technicians' work typically feed the work of managers and professionals. Consequently, if technicians are to be folded into a vertical division oflabor, the form of organization is less likely to resemble a traditional bureaucratic hierarchy than a hierarchy of occupational dominance. The resulting status structure would therefore be more akin to that of a hospital or scientific laboratory than that of a firm. Alternately, the expansion of technical work may force organizations to return to a more horizontal division of labor. In such a world, technicians would be

52

WHALLEY AND BARLEY

viewed as members of one of a series of occupations composed of experts whose knowledge and skills must be pooled with those of experts in other domains in order to accomplish a task. Technicians would then become members of multioccupational teams. Preliminary evidence suggest that such team structures characterize the informal organization of everyday work in ultrasound (Barley 1990), histology labs (Scarselletta 1992), and scientific laboratories (Barley and Bechky 1994), even though the formal structures of these organizations still presume a more hierarchical relationship. One might view current interest in collaborative forms of organizing in industry as consistent with such a development. As knowledge becomes more extensive and differentiated, it becomes more difficult for management and members of other occupations to claim greater knowledge as a prerogative for control. Such structures may also provide solutions for engineering's potential disengagement from a close linkage with management career ladders in a world where the need for technical expertise may outrun the need for managerial expertise. The third, and perhaps most likely, scenario is that different technicians' occupations will experience different futures. In the nineteenth century it would have been difficult to predict just which categories of technical and craft work would eventually become the functions of skilled craftspeople and which would be the ultimate responsibility of engineers and management. Indeed, where the boundary ultimately came to be drawn shows significant national variation (Whalley 199la). We are in a similar situation today in predicting the future of the variety of tasks and activities we currently label technicians' work. Some technicians may find themselves treated like blue- or lower white collar labor, while others enjoy considerable occupational autonomy and recognition. Evidence for such a bifurcation of technicians' work can be found in radiology and pathology where sonographers, CT technicians, and histologists are generally granted greater respect for their expertise than are x-ray technologists and technicians in blood chemistry (Barley 1990; Scarselletta, Chapter 8). A key objective of future research on technical work must be determining the conditions that decide the technician's status and role in local divisions of labor. Arthur Stinchcombe (1959) noted long ago that both organizational and occupational forms tend to exhibit characteristics peculiar to the historical conditions of their birth. On these grounds alone, there is reason to speculate that the social organization of technical work will not simply recapitulate earlier structures and cultures of work. What is at issue is how current means of organizing work will develop in a world increasingly composed of technical workers. The issue is ultimately an empirical matter whose resolution will become increasingly clear over time. Our goal has been simply to heighten awareness that the future of work may not be well served by images from the past.

2 TECHNICAL DISSONANCE: CONFLICTING PORTRAITS OF TECHNICIANS Jeffrey Keefe and Denise Potosky

Technicians are a rapidly growing occupational group of increasing importance to postindustrial society. The business press has championed technicians as the new worker elite, and politicians predict that technicians will form the core of a new postindustrial middle class. Technicians' occupations, however, are generally not well understood. The research literature that does exist offers highly conflicting portraits of technicians as postindustrial factory operatives, as modem successors to skilled craftworkers, and as emergent professionals. The research reported in this paper attempts to bring these multiple images into focus. The previous chapter, by Peter Whalley and Stephen Barley, concluded that there are several core characteristics shared by most technicians' occupations. Technicians' responsibilities are esoteric and skilled in nature, and often involve working with or on complex technologies and technical processes. Their work requires the sophisticated manipulation of symbols and data and the diagnosis of problems. Most technicians' occupations are also characterized by the acquisition of a craft version of professional knowledge (Abbott 1988, 65). As technicians' occupations are institutionalized, some elements of formal scientific, technical, or professional training are commonly required (Carnevale, Gainer, and Schulz 1990). It is this linkage of skilled practice to a formalized knowledge system that distinguishes the technician from the skilled craftworker. Although often linked to a scientific or professional knowledge system, the technicians' skilled practices often remain tacit and contextually specific, not formally "scientific" (Barley and Bechky 1994).

54

KEEFE AND POTOSKY

Technicians' occupations are also uniquely passive in their social construction. Whereas the professions and crafts are often constructed by the active social organizing of their prospective incumbents, technicians' occupations are largely constructed and defined by more powerful others. A technicians' occupation is often the by-product of the integration of a technical profession into large corporate or public enterprises. The new occupation is assembled from tasks discarded by professionals through the "hiving off' process described by Whalley and Barley. As a consequence of the hiving off process, most technicians work in complex organizations where they neither set the entry and performance standards of their occupation, nor control the educational process through which new recruits are trained. Most cannot formally self-regulate or self-govern their work practices. Most important, technicians often operate within an established profession's field of knowledge and competence (Abbott 1988). The result of performing their skilled work within an established profession's jurisdiction is a blurred identity. The allied profession is socially acknowledged to control the technicians' entire knowledge system, whether theoretical or contextual. As a result, technicians are highly skilled workers who lack any formal control over their own occupations; they work within the orbit of a dominant allied profession (Trice 1993). In the formation of technician's occupations, the process of hiving off has followed two models of development; one is organizational, the other occupational. The distinction between these models depends on whether technicians substituted for or complemented scientists or other professionals in the specific field. For example, science technicians, as a distinct occupational group, appeared at the historical moment when science was bureaucratized and relocated into large public and corporate laboratories. As substitutes in the public health laboratories after 1895, technicians performed standardized diagnostic testing so skilled bacteriologists could devote their time to research on infectious diseases (Gosse! 1992). In the testing laboratories, technicians were substituted for scientists and were thus integrated into an organizational division of labor. With the formation of industrial research laboratories, such as GE ( 1900), Du Pont (1902), and AT&T (1911 ), first-rate university-based scientists were often promised the assistance of technicians as a recruiting enticement, along with higher pay, better equipment, and more research time (Reich 1985; Hounshell and Smith 1988). By employing technicians to conduct the more routine and physical aspects of their research, scientists could devote more time to theoretical endeavors (Roberts 1983). In this setting, technicians served to support and complement the work of scientists; consequently they were integrated into an occupational division of labor. The technician in the industrial research laboratory remained closely tied to the scientist in an occupational division of labor. The scientist possessed author-

CONFLICTING PORTRAITS OFTECHNICIANS

55

ity over the technician and defined the nature, scope, and meaning of the technician's labor (Shapin 1989, 562). Technicians became largely invisible because they worked within an occupational division of labor where a profession was acknowledged to control their entire knowledge system. Both the theoretical and practical skills of technicians were within the profession's jurisdiction, which denied them an independent identity. Technicians became visible within the organization only when they made mistakes, departed from their assigned routines, or demonstrated incompetence (Shapin 1989, 558). In the first section of this paper we use Current Population Survey data to analyze, compare, and contrast technicians, technical professionals, and other technical workers to assess how close] y technicians' occupations orbit the technical professions. In the second section we describe two groups of technicians within a single company. One group works in a manufacturing organization and the other is employed in an occupational division of labor. We also compare both groups of technicians to skilled trades workers in this same company. This comparison allows us to explore how each group of these skilled technical workers attempts to control its own training, education, skills, work practices, and its relationship with management. Management efforts to increase the technization of the skilled trades by upgrading their educational requirements in this company have encountered substantial resistance. As a group, technicians have been found to be dissatisfied with their status and treatment within the firm (Roberts et al. 1972). They have also been found to be dissatisfied in terms of their autonomy, their career opportunities, and their opportunities for self-development, and as individuals. Professional career expectations formed in educational programs have rarely conformed to their job experiences. Prior research identifies status incongruity and role ambiguity as major sources of technicians' dissatisfaction and dissonance leading to widespread feelings of deprivation (Roberts et al. 1972, 36). The occupational roles of these technicians were often defined by the professionals with whom they worked. Technicians' invidious comparisons with professionals (Evan 1964) and high levels of status inconsistency were correlated with expressed dissatisfaction, increased levels of interest in unionization, heightened interest in occupational certification, and reduced occupational and organizational commitment (Koch 1977). Status conflict, role ambiguity, invidious comparisons and their associated dissatisfaction and dissonance are thought to arise because technicians often operate in conjunction with an established profession. Prior research has compared technicians' job satisfaction and status with professionals; in the last section of this chapter we use survey results to examine their job satisfaction in relation to the satisfaction of production workers and skilled tradesworkers to question how dissatisfied technicians actually are with their status and power.

56

KEEFE AND POTOSKY

In the research reported in this paper, we have used three distinct research methodologies to help focus these multiple images of technicians: archival data analysis, case study, and attitude survey. Our research effort is exploratory and intended to be hypothesis-generating.

TECHNICIANSWITHINTHETECHNICAL OCCUPATIONAL STRUCTURE: COMPARATIVE PATIERNS

By examining commonalities, differences, and changes among and within professional, craft, and operative technical occupations (using Current Population Survey data from 1983 and 1993), we uncover some characteristic patterns that help clarify the relationships, connections, and associations of technicians with other technical occupational groups. Overall, the data on employment growth, wage growth, gender composition, and unionization strongly indicate a high level of interdependence and interconnection between technicians and technical professionals. A total of four million working technicians accounted for 3 percent of US employment in 1993 (Table 2.1 ). Between 1983 and 1993, approximately one million new technician jobs were created, increasing technician employment by 31 percent, a rate 70 percent faster than overall U.S. employment growth (18 percent). Only employment of technical professionals grew at a faster pace (36 percent), which was double the rate of U.S. labor force growth during the decade 1983 to 1993. In aggregate and within more detailed occupational categories, professional employment grew faster than technician employment; however, the patterns suggest that the growth of professional and technicians' occupations are linked. Engineers and engineering technicians posted the slowest rates of employment growth (11 and 6 percent respectively), while employment of health care professionals and health technicians each grew at 37 percent. Technicians were also less numerous than technical professionals, representing only threefifths the number of technical professionals in 1983 and slightly less than threefifths in 1993. Technicians closely reflected the gender and racial composition of the overall labor force in 1993. Technicians were 51 percent female and 10 percent black in 1993, as compared to the U.S. labor force, which was 46 percent female and 10 percent black. While technical professionals also resembled the labor force in gender composition (45 percent female), the professions employed blacks (5 percent) at half the rate at which they were found in the labor force. On the other hand, the skilled blue collar occupations of construction trades (2 percent female) and mechanics and repairers (3 percent female) remained almost exclusively male domains; they have made some progress in increasing black participation

CONFLICTING PORTRAITS OFTECHNICIANS

57

TABLE 2.1 Changing composition of employment among technical occupations, 1983-1993 Technical Operations (Numbers in Thousands)

1983

1993

Percentage Change

Technical Professionals Engineers, surveyors, architects Math and computer scientists Natural scientists Health diagnosing Health treating Technicians Engineering technicians Computer programmers Science technicians Health technicians Mechanics & repairers Construction trades Precision production Total skilled technical occupations

5,130 1,675 463 357 735 1,900 3,053 822 443 202 I, Ill 4,158 4,289 3,685 20,315

6,952 1,859 1,051 531 909 2,602 4,014 870 578 261 1,522 4,416 5,004 3,758 24,144

36 II 127 49 24 37 31 6 30 29 37 6 17 2 19

Machine operators & assemblers Transportation operators Total skilled technical operators

7,744 4,201 11,945

7,415 5,004 12,419

-4 19 4

100,834

119,306

18

20% 12%

20% 10%

Total U.S. labor force Percent of labor force in: Skilled technical occupations Semi-skilled technical occupations

Note: Sources for Occupational Employment: !983: Handbook of Labor Statistics (Bulletin 2340 USDOL 1989) Table 18. 1983: Employment and Earninfis (Jan 1994, USDOL) Table 22.

(7 percent in each category), but without much improvement in the share of black employment during the last decade. Thirty years after Title VII, gender exclusion remained a core characteristic of traditional skilled trades occupations and their progeny. This record, however, does not compare unfavorably with the largest technical profession, engineering, which was only 9 percent female and 4 percent minority in 1993. Technicians joined unions in increasing numbers in the 1970s, as reflected in the growth of union coverage for engineering and science technicians from 16 percent in 1974 to 19 percent in 1980 (Kokklenberg and Sockel11985). Most of this growth occurred in the public sector. The pervasive decline of unionization in the 1980s was shared by all technical occupations (Curme, Hirsch, and MacPherson 1990). Unionization of technicians fell from 19 percent in 1983 to 12 percent in 1991, while unionization among precision production, craft, and

58

KEEFE AND POTOSKY

repair workers went from 36 percent to 26 percent in the same period. Professionals were the group least affected by declining unionization, going from 30 percent to 26 percent union coverage between 1983 and 1991. However, union coverage of technical professionals was only 13 percent in 1991. The experiences, trends, and levels of unionization of technical professionals and technicians are highly similar. Both groups have unionized at about one half the rate of skilled and semi-skilled blue-collar workers and non-technical professionals. Technical professionals experienced rapid real wage growth between 1983 and 1993. During this period, real wages increased 20 percent, with large increases for health diagnosis (51 percent) and health treating (30 percent) professions (Table 2.2). Technicians, on the other hand, experienced a modest 3 percent real wage gain for this decade, with real wage decreases for engineering ( -4 percent) and science ( -9 percent) technicians and increases for computer programmers (13 percent) and health technicians (11 percent). That is, technicians' wage growth failed to keep pace with technical professionals in any specialized occuTABLE 2.2 Median weekly earnings and real wage changes for technical occupations,

1983-1993 1983

1993

Technical Operations

Wages

Wages

Technical Professionals Engineers, surveyors, architects Math and computer scientists Natural scientists Health diagnosing Health treating Technicians Engineering technicians Computer programmers Science technicians Health technicians Mechanics & repairers Construction trades Precision production Total skilled technical occupations

$497 $594 $542 $512 $506 $393 $357 $390 $473 $369 $294 $376 $374

$818 $902 $895 $722 $994 $687 $528 $550 $747 $501 $458 $504 $495

~378

~490

$404

$595

20% 7% 20% -4% 51% 30% 3% -4% 13% -9% II% -II% -13% -IS% 2%

Machine operators & assemblers Transportation operators Total skilled technical operators

$262 $328 $286

$333 $447 $379

-18% -9% -13%

Total U.S. labor force

$313

$463

3%

Real Wage Change

Note: Sources for Occupational Employment: 1983: Handhook of Lahar Statistics (Bulletin 2340 US DOL 1989) Table 43. 1983: Employment and Earnings (Jan. 1994, USDOL) Table 56. CPI Deflator: Economic Report of the President, 1994. Table B-59.

CONFLICTING PORTRAITS OFTECHNICIANS

59

pational category. Consequently, the wage gap between technicians and technical professionals grew during the decade. Technicians earned 72 percent of the professional's weekly earnings in 1983; however, by 1993 they received only 65 percent of professional earnings. By keeping pace with inflation, however, technicians' earnings significantly out-performed those of both skilled and semiskilled blue collar technical workers, who experienced real wage declines ranging from 11 to 18 percent. In 1983, technicians earned 5 percent less than did skilled blue-collar technical workers, but by 1993 they earned 5 percent more than these workers. The 5 percent higher median pay received by technicians lends further support to the claim that technicians are an increasingly important skilled occupational group. In terms of education, technicians, more than any other occupational group, were required to possess a two-year associate's degree as a minimum qualification, according to data collected in a 1984 special supplement to the Current Population Survey. According to this supplement, 24 percent of technicians needed at least a four-year bachelor's degree and 20 percent needed, at minimum, a twoyear associate's degree. Since two-year vocational degrees were still relatively new in 1984, we believe it likely that this requirement has expanded over the last decade. In contrast to the 24 percent figure reported for technicians, almost twothirds of technical professionals were required to have earned a bachelor's or a more advanced degree in order to qualify for their present job, and 12 percent of technical professionals needed an associate's degree. While relatively few bluecollar skilled or semi-skilled workers were required to have a bachelor's degree, approximately 5 percent of the skilled blue-collar technical workforce needed an associate's degree to qualify for their positions, which was one quarter the rate required of technicians. The data on both real wage growth and qualifying education suggest that technicians may be viewed as an intermediate occupational category situated between technical professionals and skilled blue-collar workers. The growing identification of technicians with two-year associate's degrees suggests a possible institutionalization of this unique intermediate position, identifying them as members of neither a craft nor a profession. Although technicians lagged behind technical professionals in both employment and wage growth over the last decade, the patterns of growth show a high degree of association at both the aggregate and more detailed occupational levels. The more rapid employment and earnings growth of professionals does not support the notion that technicians serve as a lower-skilled and lower-waged substitute for professional labor. In aggregate and within each specialized occupational group, the expansion of professional employment and increase in earnings exceeded the growth rates for technicians from 1983 to 1993. A more likely explanation of this pattern is that technicians served to complement and support

60

KEEFE AND POTOSKY

technical professionals in the technical division of labor. This explanation suggests that the demand for technicians is derived at least in part from the demand for technical professionals. Taken together, this data can be used to construct a composite sketch of technicians as an occupational group that combines the attributes of both the skilled craftworker and the professional. Their intermediate position may reflect an occupational group in the process of institutionalizing its unique identity. This optimistic image is widely reported in the popular press. For example, one recent report in Fortune magazine described technicians as the new "worker elite" who have become the core employees of the information age. As a result of this new status, technicians' occupations have allegedly become the new anchor for people's careers (Richman 1994, 56). Yet there remains some evidence not consonant with this portrayal. Technician's wages closely matched those of skilled blue-collar workers, and the wage gap between technicians and technical professionals has widened. Most notably missing from these popular images of the new "worker elite" are two attributes central to the definition of a craft or a profession: occupational members set entry and performance standards and individual members engage in self-regulation with considerable autonomy in their work practices (Hall 1968; Freidson 1986). Neither is characteristic of technicians' occupations. Instead, technicians may be subordinated within complex divisions of labor, either organizational or occupational (Freidson 1977), without any real institutionalized power over their occupation or their work (Simpson 1985). This situation may give rise to their widely reported dissatisfaction. The macro data, however, cannot adequately explain contextual questions of occupational control and autonomy. To examine these issues we conducted to a case study at a large pharmaceutical company in New Jersey.

A CASE OF ANALYSIS COMPARING TECHNICIANS AND SKILLED TRADESWORKERS

A large pharmaceutical company is an ideal setting for the study of technicians. Chemicals and related industries, such as pharmaceuticals, employ 19 percent of the nation's quarter million science technicians. At the request of the Joint Labor-Management Apprenticeship Council, we developed a survey covering a variety of work-related attitudes, including job satisfaction, work characteristics, and preferences for different types of work within the organization. We began our project in September 1992 and completed it in August 1993. During this period we conducted over forty interviews with individuals at all levels within the organization, held several focus groups, at-

CONFLICTING PORTRAITS OFTECHNICIANS

61

tended a variety of meetings, and surveyed 337 workers, including forty-three laboratory technicians, ninety-six skilled tradesworkers, and 198 semi-skilled production workers. The information obtained in the interviews, focus groups, and meetings form the basis of our case analysis. The results of the survey are presented in the last section of this chapter. The workforce we studied was approximately 60 percent female and 30 percent minority. The technicians, skilled trades workers, and other workers were all represented by an industrial union with a single local at this location. In 1993, this facility employed 190 skilled tradesworkers in its maintenance department. These workers were members of a wide range of occupations, including the construction trades, such as electricians, carpenters, and pipe fitters; mechanics and repairers, such as millwrights, instrument repairers, refrigeration mechanics, and various specialized industrial machinery mechanics; and precision production workers, such as machinists. Another 650 workers were employed in manufacturing. Approximately 125 science technicians were employed in the Quality Control (QC) and Research and Development (R&D) laboratories at this site. Ninety of these were chemical technicians; thirty-five were biological technicians.

Diversity in Lab Work of the Quality Control and R&D Laboratories The QC labs were an integral component of the company's manufacturing organization. They were responsible for testing raw materials, drug containers, and manufactured pharmaceuticals. Under FDA regulations, the company was required to track the manufacture and testing of each batch of pharmaceuticals it made. Working from a batch number, the company's records indicated when a drug was made, when the raw materials were received, who tested the raw materials, and when they were tested. The records provided similar information for containers. Records also documented who tested the manufactured drug, what tests were performed and what results were obtained, and where and when the drug was tested. A sample from each batch of pharmaceuticals was kept in a controlled environment for a specified period of time after its manufacture. This highly integrated information system required each QC technician to follow strict standardized testing and documentation procedures. Technicians were expected to document their tests precisely. Consequently, quality control technicians performed repetitive, carefully prescribed procedures, known as "cookbook science." In the course of the last ten years, electronic automation had dramatically changed QC testing. Technicians no longer counted, weighed, or measured; machines performed these tasks, along with many others. Monitoring and maintaining the expensive electronic and testing equipment had become a major part

62

KEEFE AND POTOSKY

of a technician's job. If there were testing problems, technicians needed to be able to determine whether the trouble was in the equipment or in the quality of the material being tested. Most of their work, however, was routine and boring. For this reason, several people interviewed commented that the QC lab techs had the worst jobs on the site. The technicians who worked in R&D were at the bottom of a highly credentialed hierarchy. Scientists held doctorates; associate scientists had either M.S. or B.S. degrees, and new lab techs at a minimum held associate's degrees. R&D, unlike QC, occasionally hired individuals with bachelor's degrees into the bargaining unit to be lab techs. Although there was a highly formal status and authority hierarchy in R&D, lines of work were much less clearly defined than in QC. Work in R&D was considerably more flexible. Like lab work in QC, R&D lab work required a considerable amount of documentation, since successful projects might eventually lead to FDA drug trials. However, R&D permitted a much greater diversity of work experience for technicians than did QC. Technicians could assume more responsibility, if they wanted it. The company, however, could hold them accountable only for the practices covered in their job descriptions. While many tasks were mundane, R&D technicians might be asked to work on more interesting tasks such as instrumentation or experimental profiles. In R&D, technicians could work one-on-one with a scientist or in project teams. Most often, technicians were assigned to scientists, and, as one company representative explained, the scientists "wanted flexible people who [didn't] complain." Thus, successful work relations often depended on the ability of the technicians to get along with the scientists. Neither the company nor the union made any formal distinction between QC and R&D technicians in terms of job title (they were all "technicians") or pay level. Yet substantial differences existed between QC and R&D technicians in terms of supervision (by lab supervisors vs. by scientists), the nature of their work (equipment monitoring and troubleshooting vs. complementing the work of scientists), and the quality of the work environment (routine interface with equipment and supervisors vs. interaction within project teams or one-on-one with scientists). Many R&D technicians' jobs required skilled practice in a variable and changing environment often highly dependent on their social interactions with scientists. The QC technicians' jobs did not. Their work systems in many ways resembled those of production workers. The quality control technicians were fully subordinated and integrated into a manufacturing hierarchy which required repetitive, precise, and strictly regulated testing synchronized to the rhythm of automated production. In order to understand these differences better, it is useful to examine the training and work history of technicians relative to that of the skilled trades workers in this organization.

CONFLICTING PORTRAITS OFTECHNICIANS

63

A Case History of Apprenticeship Training for Technicians and Skilled Tradesworkers Over the last two decades, the training programs for both technicians and skilled tradesworkers underwent major restructuring, which remained a source of ongoing tension between the company and union. Controversies over the appropriate domains of occupational knowledge and skill have centered around the issues of selection criteria and educational requirements for each of these job titles. Until the mid-1970s both science technicians and skilled tradesworkers were trained in federally registered apprenticeship training programs. These four-year programs combined evening study in vocational and technical high school courses with a series of on-the-job demonstrations. In the course of the four-year apprenticeship, a lab tech or skilled trades apprentice was required to demonstrate his or her competence in 100 to 150 tasks before being certified as a "journeyman" or "technician." The apprenticeship program provided an opportunity for upward mobility for semi- skilled production workers. Whenever possible, technician and skilled trades apprentices were drawn from the production workforce. The most senior qualified bidders were selected. Prior to 1972, eligibility for apprentice training was determined by seniority, age, physical condition, and a written qualifying test. The maximum age for apprentice applicants was forty years. Candidates had to undergo a medical examination by the company doctor; and most important, they had to pass a vocational pre-test administered by the county apprentice coordinator for vocational and technical high schools. The test examined the candidates' mechanical aptitude and basic math and reading ability. The test, however, was never validated. In the early 1970s, the federal Bureau of Apprenticeship and Training (BAT) informed the company that the test would be indefensible if it were challenged in an Equal Employment Opportunity complaint. The company immediately eliminated the test and age requirements as selection criteria for apprenticeship training. This left seniority and the company physical as the only legally sanctioned criteria for selecting apprentice candidates. Yet the industrial psychology literature has demonstrated that seniority does not predict performance either in training or on the job. From one manager's perspective, the new selection criterion was reduced to whether the senior bidder had a "warm body." They reported that apprentice quality started on a downward spiral. Throughout the program, not only at school but in the workplace as well, apprentice training standards for both technicians and skilled tradesworkers were relaxed. In the vocational and technical school, no one ever failed. On the job, everyone was passed through

64

KEEFE AND POTOSKY

their required tasks, partly because the company was rapidly expanding and needed technicians and skilled tradesworkers. The federal government, however, soon forced management to recognize that the labs needed to improve the performance of their lab technicians dramatically.

Problems with the Quality of the Quality Control Lab Technicians In the mid-1970s, the company encountered some difficulties with the Food and Drug Administration (FDA), which regulates the quality standards and procedures for the manufacture of pharmaceuticals and approves new drugs. The company's problems with the FDA led to a shake-up in the management of the organization. A new vice president of quality control, a scientist with a Ph.D., was hired. An internal investigation initiated by the new vice president revealed an extremely high error rate in basic laboratory activities, such as measuring, weighing, and counting. The investigation concluded that the laboratory technicians were the primary source of the company's quality control problems. Investigators found that many quality control technicians were poorly selected and trained, irresponsible, and unmotivated. Technicians often did not understand the quality procedures they were implementing. They could not explain why they were doing specific tasks or tests. As one manager explained, management reached the consensus that the company needed better people in the labs- "it could no longer tolerate mediocrity." A management committee was established to explore alternatives for selecting and training laboratory technicians.

The Professionalization of Lab Technicians After weighing alternatives, the management committee concluded that upgrading could be best accomplished by recruiting technicians with associate's degrees. The local community colleges had established two-year programs for chemistry and biology technicians that were already producing highly qualified graduates for hospitals and other companies in the area. To implement this change, however, management would have to persuade the union to relinquish the laboratory technician apprenticeship program. Management argued that achieving a first-rate quality control operation was essential to maintaining its production at the facility. This argument was advanced at a time when pharmaceutical companies were being lured from New Jersey to Puerto Rico by extremely favorable changes in federal tax laws, a situation which made the union highly responsive to the company's concerns. After a series of quid pro quos were bargained, the union agreed to abolish the technician apprenticeship program. The change was ratified by the local membership in a contract vote. In return, the company would pay all educational costs

CONFLICTING PORTRAITS OFTECHNICIANS

65

for any union-represented worker who wanted to become a lab tech by attending an approved community college evening program. Like an apprenticeship, the program could be completed within four years. All lab tech vacancies would be posted internally. The company obligated itself to select the most senior incumbent employee, holding an appropriate associate's degree, before hiring externally. Since the abolition of the apprenticeship program, however, less than half a dozen incumbent workers have earned degrees to qualify for the more than forty lab tech vacancies. All other candidates have been hired from the colleges or from other employers. The new degree requirement effectively eliminated one upward track in the company's internal labor market for production workers. The company relied on the community college programs to screen and provide basic training for technicians. Enrollment screening at the colleges was accomplished by candidates taking a standardized basic skills test, which predicted success in the community college program. Further screening occurred through course grades. The degree programs required candidates to take classes in science theory and math, laboratory courses, and several liberal arts courses, including English. Community college programs also fostered an appetite for professional status among the technicians. The graduate technicians arrived at the company with high personal aspirations and a basic understanding of laboratory science. Supervisors and scientists were then formally responsible for ensuring that technicians were trained in their specific jobs. The community college educated lab technicians told us that their college labs and the math and English classes were useful to them on the job. The theory courses in chemistry and biology, however, were not considered relevant, since technicians were not responsible for developing research methods for the company. When we asked how they learned to do their jobs, the technicians initially recited the company's formal training procedures, which relied on the community colleges for basic education and held supervisors and scientists responsible for training lab techs on the job. With some probing, however, they indicated that other lab techs, and not the supervisors, scientists, or college instructors, had actually trained them. As management had hoped, the community college-trained lab technicians did improve quality control. These lab technicians understood the importance of precisely followed procedures and detailed documentation. Encouraged by their success, management decided to recruit people with more advanced degrees to be technicians. If an associate's degree was good, then logically, a bachelor's degree would be better; a master's better still; and a doctorate would be best. Because of the company's reputation as a good employer and the union pay scale, management was able to implement its upgrading plan. The company even recruited Ph.D.'s, mainly foreign-born scientists, into the bargaining unit. This plan proved disastrous. In Quality Control these more advanced degree-holders found their work boring, routine, and even demeaning. Once they realized that it would

66

KEEFE AND POTOSKY

take at least three to five years to get a promotion, many of them expressed their intense displeasure with everyone, and many quit. Learning from this experience, by 1993, Quality Control wanted only two-year degree technicians. Quality Control did have a limited number of professional positions within the bargaining unit, such as analytical chemist, bacteriologist, and microbiologist. Each of these job titles had two levels, one for B.S. graduates, the other for M.S. graduates. Supervisors in Quality Control held at least a bachelor's degree and were usually promoted from the professional ranks of analytical chemists, bacteriologists, and microbiologists. Since most lab techs possessed at most an associate's degree, this practice of promoting only those with more advanced degrees effectively eliminated any upward mobility for technicians and effectively denied them any formal decision making power or control over their occupation. Management's investigation of the QC labs' quality problem revealed that lab techs were not performing even basic laboratory activities, such as measuring, weighing, and counting, accurately. They attributed this problem to poor selection, training, and motivation of apprentice-trained lab techs. Once diagnosed, the problem was readily solved by seeking a new source of supply. In a labor market with an excess supply of educated labor (Freeman 1976), management's demand for community college-trained technicians was relatively easy and inexpensive to fill. Yet, did it solve the difficulty or was it, possibly, a solution looking for a problem?

The Framing of Quality Control's Quality Problem Does Not Yield a Quality Solution If the apprentice-trained lab techs failed in their basic tasks, management's logic suggested, they were incapable of performing more complex activities. However, if lab techs were not accurately measuring, weighing, and counting in the QC lab, was this behavior tolerated in production workers, the source supply of the apprentice-trained lab techs? We found no evidence to suggest that it was. One decade later this same quality problem probably would have been solved using Total Quality Management (TQM) techniques. Analogous to Frederick Taylor's 1911 analysis of productivity, TQM argues that quality is fundamentally a management problem. It identifies only two possible sources of employee mistakes: lack of knowledge and lack of attention. Lack of knowledge can be easily solved by training while lack of attention "must be corrected by the person himself or herself, through an acute reappraisal of his or her moral values" (Crosby 1985, 83). If reappraisal fails, then employee discipline and discharge remain as the only viable solution. From the TQM perspective, the QC lab techs did not suffer from lack of knowledge, but from a lack of attention. Education

CONFLICTING PORTRAITS OFTECHNICIANS

67

and training did not comprise the prescribed solution for a lack of attention. At its root cause was a moral question of whether or not lab techs were going to ensure that the manufactured pharmaceuticals conformed to established quality control standards. According to TQM, it was management's job to guarantee that their answer was conformance. Instead, management in the 1970s chose credentials over conformance. By recruiting community college-trained lab techs into this production environment to solve its quality problem, however, management engendered a new set of difficulties. In response to its quality control issues, management attempted to replace apprentice-trained technicians with college-trained technicians. This effort ordinarily might be expected to professionalize the technicians' ranks. In the QC labs, however, the new techs were integrated into an industrial bureaucracy which with the advance of technology brought about a greater routinization of QC work. Eventually, automation eliminated the potential for human errors in measuring, weighing, and counting. In R&D, the new technicians were formally slotted into the bottom of R&D's occupational hierarchy.

Technization of the Skilled Trades In the early 1980s, management once again decided to address a problem by relying on community college programs, this time for skilled tradesworkers. However, the outcome of this effort was vastly different from what happened to the technicians. The rapid diffusion of new electronic controls prompted the company to demand a change in the apprenticeship program. The journeymen who would be responsible for maintaining electronic controls and the instrument repairers were only skilled in working on pneumatic control devices. They had received no training in electronics. From management's perspective, the trades apprenticeship program was also on the same downward spiral that had caused alarm about the quality of lab tech training. In the past, management had said that they could tolerate some poorly trained journeymen, because they could be assigned routine or "scut" work. The new technology, however, was rapidly eliminating such work. Managers also noted a change in the motivations of the apprentices. Traditionally, apprentice candidates had often wanted to learn a specific trade; now, the candidates were primarily interested in making more money and did not care what trade they worked in. This indifference to one's trade, according to management, meant that these apprentices were less motivated to learn at school and on the job. The program's major problem was that it was unable to screen out these less motivated apprentices. For the traditional program to work, it needed highly motivated, self-disciplined apprentices. Managers felt that they had lost control over the selection and screening of apprentices, because of the Equal Em-

68

KEEFE AND POTOSKY

ployment Opportunity (EEO) required elimination of the test and because of a steady decline in the educational quality of vocational and technical schools. As technology advanced, the vocational and technical high schools were unable to keep up, further exacerbating management's disaffection. As one manager explained, when the company wanted to change the instrument repair apprentice program from pneumatics to electronics, the "voc-tech had nothing to offer us." A management committee was established, under the leadership of the same manager, an engineer by training, who had spearheaded the changes in the lab tech program. Management decided, once again, that the vocational and technical high school program represented a major obstacle to improving the quality of the skilled trades apprenticeship program. The company informed the union that it would not post any new apprentice openings and threatened to begin subcontracting the work unless progress was made on the school quality issue. To solve the immediate problem of getting electronic instruments training, management proposed that the course work be moved to a well-respected technical institute in the area. In 1984, the union relented, after being promised that several new apprentice job openings would be posted. At a membership meeting, the local union ratified the changes in the apprenticeship program. Shortly thereafter, management proposed modifications in school requirements for each of the other trades' programs. This time the course work moved from the voc-tech to the local community college. Again, new apprentice openings were exchanged for each trade's concession to relocate their formal training to the college. In order to attend the community college, the apprentice candidates were required to pass the New Jersey Basic Skills Test (BST). This test was designed to predict success in college. To bid on an apprentice job opening, a candidate had to have taken and passed the BST, which provided a legally sanctioned screen on job bidders. Once admitted, each trade's college program required an apprentice to take ten or eleven courses over a four-year period. One exception existed for the instrument repair apprentices, who continued to attend the technical institute. From its inception in 1984, to 1993, the new apprenticeship program graduated eighteen journeymen. In 1993, six apprentices were enrolled. Six other apprentice candidates had left the program before graduation: three were "kicked out" when they failed college courses, two bid out to non-trades jobs because they were not passing their task requirements, and one was bumped out in a layoff. The new program had a 25 percent attrition rate even after using the Basic Skills Test as an admission screen. Management believed that the new program had solved both the apprentice quality issue and advanced technology education problems. This point of view was not shared by the skilled trades, who unanimously opposed the new program. In order to understand the skilled trades' opposition to the community college based formal education, it is necessary to consider the program it replaced.

CONFLICTING PORTRAITS OFTECHNICIANS

69

Why the Skilled Trades Opposed the Community College Program In the voc-tech program, experienced tradespeople taught virtually all the courses. These teachers could easily relate the course material to work practices. Classes met twice a week, one night in a classroom, the other in a workshop. Apprentices were encouraged to speak to their instructors about any technical problems they were having on the job. The school programs also helped socialize the apprentices into their trade. The programs enrolled students from a wide variety of workplaces. Apprentices quickly learned that their new trade provided them with a multitude of employment opportunities and could guarantee their security and mobility. On the job, apprentices were assigned to journeymen; they often started as helpers. As an apprentice gained experience, the journeyman trained him or her in a variety of tasks. When the apprentice was ready, a supervisor, who was a former skilled tradesworker, evaluated the apprentice's performance and certified the task's satisfactory completion. Trades people believed that the only legitimate test for the apprentice was whether he or she could do the job. Consequently, they believed that formal course work should be aimed only at improving job performance. The journeymen graduates of the voc-tech program thought their education had helped them immeasurably to learn on the job. However, they did not have the same opinion about the college program. The journeymen thought that most of what was taught at the college was irrelevant to a tradesperson. Several of them argued that the college program was really a barrier to training apprentices, because apprentices spent too much time worrying about their college courses, which only distracted them from mastering their trade. Even the most successful graduates of the new apprenticeship program agreed with the other journeymen that a considerable part of the college program was irrelevant. As with the lab techs, the theory courses were considered the most difficult and most irrelevant classes in the program. These courses were taught primarily to engineering students, and the instructors often assumed that everyone in the class was an aspiring engineer. Most of the apprentices' instructors were either engineers or scientists. None of the instructors or other students were skilled tradespeople. No one in the community college courses had ever worked in the trades. Several apprentices felt that the instructors condescended to them, once they learned that the apprentices were not engineering students. African American production workers also expressed dismay about the college program. Although the skilled trades were largely a white male domain (4 percent African American and 1 percent female), African American workers thought minorities would have greater difficulty than whites with the college pro-

70

KEEFE AND POTOSKY

gram. On average, the minority workforce had had less schooling, and had often attended the worst schools. Consequently, fewer minorities could qualify to bid for the apprentice openings. The African American workers believed that many minority workers who could do skilled trades work would be screened out by the college requirements. Since the new program had begun, only one African American worker had graduated, an individual who everyone agreed was of exceptional ability and tenacity. The minority workers also thought any fair evaluation of an apprentice should be based only on his or her job performance; grades in the school program were irrelevant. School was only considered useful when it supported workplace learning and achievement. The skilled trades supervisors were also critical of the new program. They, too, found that the apprentices were preoccupied with their college courses. They added that the college failed to provide apprentices with training in some basic skills, such as blueprint reading, which had been taught in the voc-tech program. According to the tradesworkers, the move to the college to improve the "quality" of journeymen implied that management believed that they were inferior tradespeople; a conclusion with which they disagreed and which made many of them angry. The union local president, a skilled trades journeyman, believed that the original resolution of the instrument repair training problem was a mistake. In every venue we observed the tradespeople voicing their objections to the college program. They fought vigorously to get the program moved back to the voc-tech and placed under the control of educators who were friendly to the tradesworkers' knowledge system. Management, on the other hand, resisted the skilled trades' demand to relocate the course work. They pointed to the building trades, who had largely moved their apprenticeship programs out of the voc-tech program, as a positive example of the new system. The building trades, however, still retained control over their programs, whether they were located in their own facilities, in community colleges, or in voc-tech high schools. In addition, their programs had sufficiently large enrollments for them to hire their own instructors and exert influence over course content and teaching. In a recent evaluation of the apprenticeship program, management had hired an industrial psychologist to validate the training that the apprentices received in college. It is likely that management assumed the consultant would support the new college-based program. However, the psychologist found that on average, two of the ten courses in each program could not be defended as job-relevant if challenged in an EEO complaint. As a result, courses were changed, and a blueprint-reading course at the voc-tech, along with one or two other courses, were added. In one respect, this change represented a union success; they had succeeded in influencing the nature of apprenticeship training for the trades. At best,

CONFLICTING PORTRAITS OFTECHNICIANS

71

this solution was a compromise, as both sides remained committed to their knowledge systems and their respective educational programs for training apprentices.

The Contest for Control over Knowledge Systems Lab techs had lost their limited control over their knowledge system when the apprenticeship program was abolished. The new associate's degree requirement not only deprived lab technicians of control over their formal education, it disrupted their workplace training system. The college programs, taught by scientists, emphasized theoretical training over practice-based learning. Technicians, however, learned diluted science. In the workplace, peer-trained task competencies had been replaced with a formal system which placed the responsibility for training in the hands of supervisors and scientists, who had never done lab technicians' work. Although some technicians, particularly in R&D, acquired a crafttype knowledge through their lab practice, many of them were not cognizant of their own knowledge system and, more importantly, it was not institutionalized. Instead, their knowledge was deprecated and, in R&D, subsumed within the larger scientific knowledge system. Since they lacked control over their own knowledge, lab tech work was largely shaped by the demands of others. In R&D it was shaped by the demands of scientific research, and in QC it was shaped by the logic of manufacturing. Lab techs in both R&D and QC were supervised by those who had never done their work, but who held higher credentials. Consequently, their upward mobility was limited. Without further education they were restricted to moving to another technician position. While this might cause them to appear like professionals, the lab technicians were in fact the bottom of the scientific status and authority hierarchy. In the QC labs, their work was integrated into pharmaceutical manufacturing and thus resembled factory work. They needed to conform to the highly interdependent demands of advanced manufacturing and information technologies. The least skilled work of the QC technician-measuring, counting, and weighing-had been automated. QC lab techs complained of highly routinized and boring tasks which were periodically punctuated with quality or machine errors. We found support for an observation, made over two decades ago, that the division of labor leads to boring and repetitive tasks for an increasing number of jobs performed by laboratory and quality-control technicians (Roberts et a!. 1972). In the R&D labs, technicians worked for scientists and they had to learn to be flexible, to adapt their practice to the scientists' idiosyncrasies, and to get along

72

KEEFE AND POTOSKY

without complaining. Even those who took initiative and assumed extra responsibilities could only hope to "assist" a scientist in the development of new methods. Although they believed that they did all of the work, most of this work was unacknowledged, unappreciated, and invisible. They were nonprofessionals living within a professional knowledge system. However, their work was often more independent and less tightly controlled, and their practices more diverse and skilled, than those of their counterparts in the QC labs. In contrast, the skilled trades workers were at the apex of the blue-collar status hierarchy. Although embattled with an engineering reconception of their occupation, they retained control over their shop floor knowledge system and maintained autonomy in their work practices. Nevertheless, they had lost some control over the formal education and socialization of apprentices. Yet, because their knowledge system was institutionalized, making them a cohesive social force, they had been able to mount a vigorous campaign through the union against the new educational system, gaining some unlikely allies. They were supervised by former trades workers, who were themselves products of the trades' knowledge system. The tradesworkers remained independent and largely defined the nature of their maintenance work, often working without any direct supervision. Journeymen were still responsible for the workplace education of apprentices. Apprentices and journeymen were also joined in a common cause to eliminate the college program and reclaim control over their craft knowledge system. The differences between the skilled trades workers and the lab technicians suggest another dimension of dissonance for the lab techs. While both groups of workers were technically skilled, neither enjoyed professional prestige; however, their social experiences were markedly different. Tradesworkers knew their interests; they were organized, and they were consciously engaged in a struggle with management to maintain control over their knowledge system. They fundamentally appreciated that an autonomous, independent, skilled practice depended on their collective action. The lab technicians, however, were neither so cognizant nor so organized. Their identity as educated, technically competent employees with professional aspirations corresponded neither with the routine structure and nature of their work in the QC lab, nor with their inferior social status in R&D. Yet the lab technicians were not prepared, organizationally or ideologically, to take collective action to gain control. They also did not view the union as a means of support for such a struggle. As a result, their status was ambiguous and largely defined by what they were not. They did not possess professional prestige, autonomy, independence, authority or pay; however, they desired it. They also did not share craft organization, autonomy, or control, nor did they accept the production workers' subordination to managerial authority. Consequently, they persisted in a state of chronic personal dissatisfaction.

CONFLICTING PORTRAITS OFTECHNICIANS

73

Lab Technicians on Management and the Union Technicians at this facility worked within either the R&D occupational hierarchy or the QC manufacturing bureaucracy; each division organized the work of technicians in markedly different ways. However, R&D and QC lab techs shared similar perceptions about themselves as technicians and their experiences with others. Many technicians, both QC and R&D, thought that they were notrespected by supervisors, scientists, or upper management. Technicians felt that those in the upper echelons looked down upon them, saying, "They're not suppose[d) to socialize with us." Technicians reported, "We do all the work," and "Scientists develop new methods, and sometimes we get to 'assist' them." One lab tech related the following story: "A director wrote an official memo to upper management about how dumb and unmotivated we are. He referred to us as 'lemmings.' Do you know what that is? Well, I didn't, so I looked it up-it's a little brown rodent!" A copy of this memo circulated among the lab technicians, and was a source of considerable irritation. Their consistently negative views toward management, however, did not translate into support for the union. Lab techs were highly critical of the union, telling us, "The union never defended us. They don't care about us," "The union is a puppet of management," "There is no union; there's only dues!" and complaining that the leaders "are all in the skilled trades." Many lab techs thought the union was run by and for the skilled trades, which was not completely accurate. The local leadership was a coalition government, drawn from two constituent groups: the skilled trades and the African American workers, who were primarily based in production. Both of these groups shared similarly hostile views about the changes in the skilled trades apprenticeship program; their leaders were surprised to learn that technicians had any problems.

WORKERS' SELF-PERCEPTION OF SKILL AND JOB SATISFACTION

We developed the survey instrument in discussions with the Joint LaborManagement Apprenticeship Council. As the council sought consensus on the inclusion of each question, some important but politically sensitive items about supervisors, management, and the union were deleted from the final survey. As noted earlier, our survey results were based on the responses of 337 workers, including forty-three laboratory technicians, ninety-six skilled tradesworkers, and 198 semi-skilled (mainly production) workers. The respondents voluntarily filled out the written employee survey on company time. Table 2.3 compares self-perceptions regarding the skilled tradespeople's, lab techs', and production workers' satisfaction, skill, and job security. It also reports

74

KEEFE AND POTOSKY

TABLE 2.3 Occupational comparisons (Five-point scale: 5 [Strongly Agree] to I [Strongly Disagree]) Occupational Comparisons of Self-Perception Variables and Scales

Occupations

Skilled trades

Are trades & lah techs' differences significant?

Laboratory technicians

Are lab techs' & production workers differences significant?

Satisfaction Skill Autonomy Complexity Market value of skill in 5 years Computer knowhow Job security

3.92 4.21 4.16 4.24 4.17

yesc yesc yesc yes" yesc

2.97 3.68 3.07 3.94 3.14

yes" no no yes" yesh

3.34 3.46 3.14 3.60 2.61

2.74

yesc

2.98

no

2.82

3.51

yes

2.98

no

2.82

Female (I ,0) White (1,0) Age (years) Education (1-9) Job tenure (years) Seniority (years) Married (I ,0)

0.02 0.94 46.60 5.90 14.86 20.39 0.85

Production workers

Occupational comparisons of demographicvariables yesc yesc yes" yesc yesb no yesc

0.44 0.63 42.04 6.66 10.38 16.95 0.56

yesc yes" yesa yesc no no no

0.69 0.45 45.43 4.51 8.15 15.80 0.56

Note: T-test significance levels for differences not zero are: "p < .05, bp < .001, cp < .0001

demographic comparisons by occupation. We examined whether the differences reported by each occupational group were statistically significant using standard t-tests. Where the differences between the responses of skilled tradespeople and lab techs or between production workers and lab techs were significant, such is identified with a "yes," and the level of significance is also indicated. If the difference between the groups' responses was not statistically different from zero, a "no" was recorded. Satisfaction, skill, complexity, and autonomy were measured on scales that we constructed using multiple items. Each scale's reliability coefficient (Cronbach 's alpha) and respective items are reported in Table 2.4. Skill was measured as a two-dimensional construct, capturing both substantive complexity and autonomy (Spenner 1990). As in prior research, these two dimensions were highly correlated. When the ten items that measure complexity and autonomy were

CONFLICTING PORTRAITS OF TECHNICIANS

75

combined into a single skill scale, the scale had a higher degree of internal reliability than did either of the two subscales (alpha = .90, as compared to .86 and .71, respectively). Job satisfaction was measured by a four-item scale with a highly acceptable level of internal reliability (alpha = .80). Returning to Table 2.3, lab techs were the most dissatisfied occupational group we surveyed. The self-report data indicated that most lab techs rated their jobs as having skill requirements similar to those of production workers. Nevertheless, their jobs were substantively more complex than the report the production workers gave of their jobs; yet lab tech jobs ranked the lowest on autonomy TABLE 2.4 Satisfaction. skill, and education scales (Five-point scale: 5 [Strongly agree] to I [Strongly disagree]) Satisfaction (Cronbach's Alpha • • • •

=

.80)

I am satisfied with the work I do at the Company. My work is interesting and challenging. I am satisfied with my opportunities for advancement. I intend to stay in my present job for as long as possible.

Skill (Cronbach's Alpha

=

.90)

=

Complexity + autonomy (10 item scale)

Complexity (Cronbach 's Alpha = .86) • My job requires me to use my memory. • My job requires accuracy and precision. • My job requires knowledge of math. • My job requires reading and spelling. • My job requires the ability to clearly communicate. • My job requires eye and hand coordination. • My job requires concentration. Autonomy (Cronbach's Alpha= .71) • I have the opportunity to exercise my own judgment on the job. My job requires me to regularly solve problems. • I use all my skills from my experience and training on my present job. Education Please specify the highest level of education you have completed (circle number). I. Some grade school 2. Completed grade school 3. Some high school 4. Completed high school 5. Vocational training school 6. Apprenticeship program 7. Some college (includes 2-year associate's degree) 8. Completed college 9. Some graduate work

76

KEEFE AND POTOSKY

(though not significantly different from the production workers.) In comparison with the production workers, lab techs thought that their skills would be more valuable in the market in five years, but they did not necessarily feel secure in their current jobs. The lab techs ranked significantly below the skilled tradesworkers in terms of job satisfaction, skill, and job security. On only one item, knowing how to use a computer, did the lab techs report greater competence than both the trades and semi-skilled workers. The lab techs were also the youngest and best-educated group we surveyed. They were significantly more racially integrated than the skilled tradesworkers. Lab tech employment of women reflected the overall labor force's proportion of female employment, which was in sharp contrast to the skilled trades. On the other hand, the production workforce we surveyed was mainly nonwhite and female. All groups of workers exhibited long tenure in their current jobs, but the skilled trades had significantly more time in their jobs than did either the lab techs or the production workers. There was no significant difference, however, among the groups in company seniority. Skilled trades workers were significantly more likely to be married than either the lab techs or the semi-skilled workers. When we examined the individual items that comprise the scales, we found much of the lab techs' dissatisfaction to be associated with their lack of autonomy, their routinized work, the dearth of formal opportunities to exercise judgment and solve problems, their inability to use their skills on the job, insufficient opportunities for advancement, and lack of acceptance by their co-workers. On each of these items, their attitudes more closely matched those of production workers and were significantly more negative than those held by the skilled tradesworkers. Yet in required job skills, such as math, reading, accuracy, and precision, their self-perceptions more closely approximated those of the skilled tradespeople rather than those of the production workers. Table 2.5 reports the coefficients from three linear models that predict employee job satisfaction, estimated using a population weighted least squares. Model I reveals the importance of skill in predicting satisfaction. Once skill and the future market value of skill were held constant, the difference between production workers (the omitted category) and the skilled tradespeople in job satisfaction became statistically insignificant, suggesting that the skilled tradespeople were production workers with more skill. On the other hand, lab techs remained significantly less satisfied. Even when we control for perceptions of job security, computer use, and demographic differences in models 2 and 3, lab techs remained significantly less satisfied than either the production workers or the skilled tradespeople. These results suggest that we need to look at factors other than skill,job security, or differences in worker backgrounds to explain the relative dissatisfaction of the lab techs.

CONFLICTING PORTRAITS OFTECHNICIANS

77

TABLE 2.5 Predictors of job satisfaction Dependent variable: Added satisfaction scale Mean = 13.24 Skill Market value of skill

Modell 0.22c (.02) 1.02C (.15)

Job security Female Seniority

Model2

Model3

O.J8C (.03) 0.6SC (.17)

O.J9C (.03) 0.6JC (.17)

o.nc

0.7oc

(.15) 0.96b (.37) 0.04" (.02)

(.16) 0.92b (.38) 0.03 (.02) -0.01 (.02) -0.19 (.16) -0.12 (.37) -0.21 (.36) -0.12 (.13) 0.61 (.65) -1.34b (.53) 3.J6b (1.13) .48

Job tenure Computer White Married Education Skilled trades Lab technician Intercept Adjusted R-squared

-0.65 (.51) -2.1SC (.43) 2.89b (.81) .44

.26 (.56) -J.79C (.43) 1.89" (.84) .50

Note: T-test significance levels for beta not zero are: ap < .05, bp